PDA

View Full Version : Could Western Jews (Ash. and Seph.) descend from Aegeans and Levantine admixture?



Pages : [1] 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43

Erikl86
06-05-2018, 09:54 PM
I've recently tried to find the best fit for Ashkenazi and Sephardic Jews on nMonte, using Eurogenes Basal-rich K7 data.

These are the results I got:

[1] "distance%=0.0677 / distance=0.000677"

Ashkenazi_Jew:average

Minoan_Lasithi:I0070 42.35
Iran_ChL:I1661 22.05
French_East:French24434 20.60
Minoan_Odigitria:I9131 7.85
Mycenaean:I9041 4.00
Levant_N:I1704 2.40
Iran_ChL:I1665 0.65

Overall Bronze Age Levant - ~25%.
Ancient Aegeans - ~54%
Eastern French (Rhineland French) - 20.6%




[1] "distance%=0.0135 / distance=0.000135"

Sephardic_Jew:average

Mycenaean:I9006 10.20
Minoan_Lasithi:I0074 9.05
Iran_ChL:I1665 9.05
French_South:SouthFrench3326 8.60
Mycenaean:I9010 8.50
Iran_ChL:I1661 8.40
Minoan_Lasithi:I0071 7.90
Iran_ChL:I1662 7.25
French_South:SouthFrench3947 5.90
Minoan_Odigitria:I9130 5.15
Levant_N:I0867 5.10
Levant_N:I1704 4.40
Minoan_Odigitria:I9131 4.25
Minoan_Lasithi:I0073 3.60
Mycenaean:I9041 2.65

Overall Bronze Age Levantine - ~34% (10% higher than Ashkenazi Jews).
Ancient Aegeans - ~51%.
South French (very close genetically to North Eastern Spanish people like Catalans) - 14.5%.

These are the closest fits I got to both Ashkenazi Jews and Sephardic Jews. Neolithic Levantines + CHL Iranians are basically Bronze Age Levantines, based on Haber et al. 2017 study:

We computed the statistic f4(Levant_N,
Sidon_BA; Ancient Eurasian, Chimpanzee) and found populations from the Caucasus and Iran shared
more alleles with Sidon_BA than with Neolithic Levant (Figure 2A). We then used qpAdm8
(with parameter allsnps: YES) to test if Sidon_BA can be modelled as mixture of Levant_N and any other
ancient population in the dataset and found good support for the model of Sidon_BA being a mixture
of Levant_N (48.4± 4.2%) and Iran_ChL (51.6± 4.2%) (Figure 2B; Table S3).

From the look of it, I can't but come to the conclusions that both communities, Ashkenazim and Sephardim, are the descendants of Hellenistic Jews, which were basically a mixture of Southern, mainly Greek and Greco-Roman converts with Levantine Jews, centuries before the destruction of the second temple occurred.

According to historical sources, up to 10% of the Roman empire was Jewish, and in the Eastern half districts - up to 25% (!) of the population was Jewish.

To understand the extent of how popular Judaism became among Greeks (especially women, as the pain of circumcision and the social ramifications of having no foreskin seem to have deterred many men from fully converting) since the 2nd century BC, culminating in the 1st century AD, here’s what wikipedia has to say about the issue:


As Jews emigrated and settled in the Roman provinces of the Empire, Judaism became an appealing religion to a large number of Pagans, for many reasons; God-fearers and proselytes that underwent full conversion were Greeks or Romans, and came from all social classes: they were mostly women and freedmen (liberti), but there were also artisans, soldiers and few people of high status, like patricians and senators.

And one of the reasons why mainly women converted:


The rite of circumcision was especially unpopular in Classical civilization because it was the custom to spend an hour a day or so exercising in the nude in the gymnasium, and males did not want to be seen in public deprived of their foreskins. Hellenistic and Roman culture found circumcision to be cruel and repulsive.

And this is another source:


During the Hellenistic and early Roman period (323 B.C. to 70 A.D.), the Jews were apparently extraordinarily successful in winning converts. Natural increase alone can hardly account for the vast growth in Jewish population, since there is no evidence that the population of the world at large had increased significantly during this period or that health conditions had improved or that Jews had previously been practicing birth control. On the basis of Biblical and archaeological data, Salo estimates that Judea, which contained the major part of the Jewish population at the time of the destruction of the First Temple in 586 B.C., had no more than 150,000 Jews. By the middle of the first century A.D. he estimates(8) that the world Jewish population had risen to about eight million. In the Roman empire, he suggests, every tenth inhabitant was a Jew. The most likely explanation of this increase is proselytism, as alluded to by numerous references in Philo, Josephus, the New Testament, Greek and Roman writers and the Talmud.(9) One may cite as samples the statements that “We are much more numerous, and like the Jews we shall force you to join our throng” (Horace, Satires 1.4.142–143), and “The Holy One, blessed be He, sent Israel into exile among the nations only for the purpose of acquiring converts” (Talmud, Pesahim 87b).(10)

I believe this makes a strong historical case that the Mediterranean admixture of Western Jews (Ashkenazi and Sephardic) is with ancient Aegean people, rather than Italians, and happened prior to the destruction of the Jewish temple in 70 AD.

However, the strangest thing is the closeness to Cretan Minoans rather than Mycaneans. I've asked myself where could Western Jews could have gotten such a high Minoan admixture. Some of it is probably due to nMonte difficulty to distinguish between Mycenaean and Minoan when the two populations are mixed, but some of it might be due to perhaps Minoan origins. Then it hit me - the Philistines most probable place of origin was Crete, although their culture Mycenaean.

Perhaps the admixture between Judeans and ancient Aegean people happened in three phases:

1. Earlier one - from the beginning of the interactions between the people of the southern Kingdom of Judah with the Philistines, which were from Crete, the origin of the Minoans (but themselves were Mycenaean-influenced because they have arrived long after the Minoans have been already absorbed by the Mycenaeans). This happened from around the 11th century BC all the way to the 6th century BC.

2. Middle one - more substantial than the earlier one - from after the Babylonian diaspora, when Philistines have disappeared as separate people - that happened around the 5th-6th centuries BC. They have probably been absorbed with the remaining southern Judeans.

3. A late one - much more substantial than the other two - since the earliest days of Hellenism around the 4th century BC, culminating in Hellenistic, Greek speaking Jewish communities all over the Levant, Egypt, Asia minor, Greek Islands, and southern Italy. Most Greeks that would settle in the Levant during the 4th century BC would still be mostly Mycenaean genetically.

After that, there was probably also a little bit of admixture with non-Greek Italic people (Latin Romans basically) and then both communities started out early in the Early Medieval Ages from Jews immigrating north from Italy to France, with the split happening by the Muslim conquest of the Iberian peninsula - Sephardic Jews are the Jews which lived in south West France, Ashkenazi Jews are the Jews which lived in Eastern France. Their main non-Mediterranean European component seem to be of French origins.

There is one issue which I've yet managed to work around, and is the timing of when the admixture with southern Europeans happened.

According to a study from 2016, the timing of the admixture with southern Europeans, mainly southern Italians, between 35-55 generations ago (500-1000 AD), and this their conclusions:


What is perhaps surprising is the timing of the Southern European admixture to ≈31-52 generations
ago, since Jews are known to have resided in Italy already since antiquity. This result would then imply
no gene flow between Jews and local Italian populations until the turn of the millennium, either due to
endogamy, or because the group that eventually gave rise to contemporary Ashkenazi Jews did not
reside in Southern Europe until that time. More detailed or alternative interpretations are left for future
studies.

What if the solution is not that the Jews didn't mix with local Italians, but that until 31-52 generations ago, there was little Greek gene flow into the rest of southern Italians?

Until 640-750 AD, most southern Italy and Sicily were still controlled by Greek speaking Byzantine Empire, and Greek was still spoken by the majority of people in some southern areas in Italy and Sicily all the way to the 11th century.

After the Lombards conquered the rest of Italy, and then were absorbed by the Frankish empire in the 9th century, the area began a process of Latinization, which while was occurring before in small amounts, accelerated immensely after Southern Italy and Sicily were completely disconnected from the Byzantine Empire.

That would mean those Greek people were then absorbed into general Latin-speaking population. But that was only the start of a gradual absorption of Italian-Greeks into the rest of the population - after the Muslim conquest of Sicily and continuous raids by Muslims on continental Italy, many Christians (of mainly Greek ancestry, as was the case for southern Italy and Sicily) fled to mainland Italy and further north.

Sicily itself took more than 3 decades to be conquered by Muslims.

I believe this integration and movement of southern, mainly Greek background, Italians with the rest of non-Greek Italians, actually took place in that time frame - 500-1000 AD. Before that, while of course there was some exchange of population, Greek was an official language of the Roman Empire (even after the split, it was still official in both halves of the empire) and so Greek speaking people could have kept their identity as such and had no pressure at all to speak any other language (all official Roman positions dictated that you'd know both Greek and Latin).

Perhaps then, "This result would then imply no gene flow between Jews and local Italian populations until the turn of the millennium, either due to endogamy, or because the group that eventually gave rise to contemporary Ashkenazi Jews did not reside in Southern Europe until that time." actually means that Greek and Phoenician admixture with Italians didn't occur in large amounts until the turn of the 1st millennium AD, thus creating the illusion that Jews didn't mix with local population there - they did, it's just that their European admixture was Mycenaean-Greek, and that admixture didn't occur among Italians until after the absorption of the Magnea Grecia by Latin-Italy around the turn of the millennium.

Agamemnon
06-05-2018, 10:25 PM
"According to historical sources, up to 10% of the Roman empire was Jewish"

Which historical sources exactly? I keep seeing this claim, but nobody ever shows proof of this.

As for your scenario, while it isn't impossible, it certainly is unlikely. Overfitting is obviously an issue in your model here.

Eihwaz
06-05-2018, 10:36 PM
That is literally the most overfitted model I've ever seen, with so many interrelated populations...

vettor
06-05-2018, 11:34 PM
"According to historical sources, up to 10% of the Roman empire was Jewish"

Which historical sources exactly? I keep seeing this claim, but nobody ever shows proof of this.

As for your scenario, while it isn't impossible, it certainly is unlikely. Overfitting is obviously an issue in your model here.

Thats a ver high %.....considering the empire ruked about 13M

Maybe its a ficticous religous number since there was only jews and "pagans" at that tlme...pre-christians

Tz85
06-06-2018, 01:50 AM
I've recently tried to find the best fit for Ashkenazi and Sephardic Jews on nMonte, using Eurogenes Basal-rich K7 data.

These are the results I got:

[1] "distance%=0.0677 / distance=0.000677"

Ashkenazi_Jew:average

Minoan_Lasithi:I0070 42.35
Iran_ChL:I1661 22.05
French_East:French24434 20.60
Minoan_Odigitria:I9131 7.85
Mycenaean:I9041 4.00
Levant_N:I1704 2.40
Iran_ChL:I1665 0.65

Overall Bronze Age Levant - ~25%.
Ancient Aegeans - ~54%
Eastern French (Rhineland French) - 20.6%




[1] "distance%=0.0135 / distance=0.000135"

Sephardic_Jew:average

Mycenaean:I9006 10.20
Minoan_Lasithi:I0074 9.05
Iran_ChL:I1665 9.05
French_South:SouthFrench3326 8.60
Mycenaean:I9010 8.50
Iran_ChL:I1661 8.40
Minoan_Lasithi:I0071 7.90
Iran_ChL:I1662 7.25
French_South:SouthFrench3947 5.90
Minoan_Odigitria:I9130 5.15
Levant_N:I0867 5.10
Levant_N:I1704 4.40
Minoan_Odigitria:I9131 4.25
Minoan_Lasithi:I0073 3.60
Mycenaean:I9041 2.65

Overall Bronze Age Levantine - ~34% (10% higher than Ashkenazi Jews).
Ancient Aegeans - ~51%.
South French (very close genetically to North Eastern Spanish people like Catalans) - 14.5%.

These are the closest fits I got to both Ashkenazi Jews and Sephardic Jews. Neolithic Levantines + CHL Iranians are basically Bronze Age Levantines, based on Haber et al. 2017 study:


From the look of it, I can't but come to the conclusions that both communities, Ashkenazim and Sephardim, are the descendants of Hellenistic Jews, which were basically a mixture of Southern, mainly Greek and Greco-Roman converts with Levantine Jews, centuries before the destruction of the second temple occurred.

According to historical sources, up to 10% of the Roman empire was Jewish, and in the Eastern half districts - up to 25% (!) of the population was Jewish.

To understand the extent of how popular Judaism became among Greeks (especially women, as the pain of circumcision and the social ramifications of having no foreskin seem to have deterred many men from fully converting) since the 2nd century BC, culminating in the 1st century AD, here’s what wikipedia has to say about the issue:



And one of the reasons why mainly women converted:



And this is another source:



I believe this makes a strong historical case that the Mediterranean admixture of Western Jews (Ashkenazi and Sephardic) is with ancient Aegean people, rather than Italians, and happened prior to the destruction of the Jewish temple in 70 AD.

However, the strangest thing is the closeness to Cretan Minoans rather than Mycaneans. I've asked myself where could Western Jews could have gotten such a high Minoan admixture. Some of it is probably due to nMonte difficulty to distinguish between Mycenaean and Minoan when the two populations are mixed, but some of it might be due to perhaps Minoan origins. Then it hit me - the Philistines most probable place of origin was Crete, although their culture Mycenaean.

Perhaps the admixture between Judeans and ancient Aegean people happened in three phases:

1. Earlier one - from the beginning of the interactions between the people of the southern Kingdom of Judah with the Philistines, which were from Crete, the origin of the Minoans (but themselves were Mycenaean-influenced because they have arrived long after the Minoans have been already absorbed by the Mycenaeans). This happened from around the 11th century BC all the way to the 6th century BC.

2. Middle one - more substantial than the earlier one - from after the Babylonian diaspora, when Philistines have disappeared as separate people - that happened around the 5th-6th centuries BC. They have probably been absorbed with the remaining southern Judeans.

3. A late one - much more substantial than the other two - since the earliest days of Hellenism around the 4th century BC, culminating in Hellenistic, Greek speaking Jewish communities all over the Levant, Egypt, Asia minor, Greek Islands, and southern Italy. Most Greeks that would settle in the Levant during the 4th century BC would still be mostly Mycenaean genetically.

After that, there was probably also a little bit of admixture with non-Greek Italic people (Latin Romans basically) and then both communities started out early in the Early Medieval Ages from Jews immigrating north from Italy to France, with the split happening by the Muslim conquest of the Iberian peninsula - Sephardic Jews are the Jews which lived in south West France, Ashkenazi Jews are the Jews which lived in Eastern France. Their main non-Mediterranean European component seem to be of French origins.

There is one issue which I've yet managed to work around, and is the timing of when the admixture with southern Europeans happened.

According to a study from 2016, the timing of the admixture with southern Europeans, mainly southern Italians, between 35-55 generations ago (500-1000 AD), and this their conclusions:



What if the solution is not that the Jews didn't mix with local Italians, but that until 31-52 generations ago, there was little Greek gene flow into the rest of southern Italians?

Until 640-750 AD, most southern Italy and Sicily were still controlled by Greek speaking Byzantine Empire, and Greek was still spoken by the majority of people in some southern areas in Italy and Sicily all the way to the 11th century.

After the Lombards conquered the rest of Italy, and then were absorbed by the Frankish empire in the 9th century, the area began a process of Latinization, which while was occurring before in small amounts, accelerated immensely after Southern Italy and Sicily were completely disconnected from the Byzantine Empire.

That would mean those Greek people were then absorbed into general Latin-speaking population. But that was only the start of a gradual absorption of Italian-Greeks into the rest of the population - after the Muslim conquest of Sicily and continuous raids by Muslims on continental Italy, many Christians (of mainly Greek ancestry, as was the case for southern Italy and Sicily) fled to mainland Italy and further north.

Sicily itself took more than 3 decades to be conquered by Muslims.

I believe this integration and movement of southern, mainly Greek background, Italians with the rest of non-Greek Italians, actually took place in that time frame - 500-1000 AD. Before that, while of course there was some exchange of population, Greek was an official language of the Roman Empire (even after the split, it was still official in both halves of the empire) and so Greek speaking people could have kept their identity as such and had no pressure at all to speak any other language (all official Roman positions dictated that you'd know both Greek and Latin).

Perhaps then, "This result would then imply no gene flow between Jews and local Italian populations until the turn of the millennium, either due to endogamy, or because the group that eventually gave rise to contemporary Ashkenazi Jews did not reside in Southern Europe until that time." actually means that Greek and Phoenician admixture with Italians didn't occur in large amounts until the turn of the 1st millennium AD, thus creating the illusion that Jews didn't mix with local population there - they did, it's just that their European admixture was Mycenaean-Greek, and that admixture didn't occur among Italians until after the absorption of the Magnea Grecia by Latin-Italy around the turn of the millennium.

Jews were dwelling in Southern Italy pre 1st century, and have the oldest Jewish community outside of West Asia. I don't think the Aegean people have a stronger case at all. Imo

kingjohn
06-06-2018, 03:50 AM
i also think the southern european admixture that western jews have is more
italian than greek :)
i saw some living dna results of full aschenazi jews they dont score aegean in the complete mode
but they do score south italy , tuscany, and some north italy
about the non -southern european ancestery in ashkenazim on this i agree with you more than likely french.....

Erikl86
06-06-2018, 04:09 AM
"According to historical sources, up to 10% of the Roman empire was Jewish"

Which historical sources exactly? I keep seeing this claim, but nobody ever shows proof of this.

As for your scenario, while it isn't impossible, it certainly is unlikely. Overfitting is obviously an issue in your model here.


If I switch from nMonte2, to nMonte3, and for example add Polish to Ashkenazi Jews, I get the following:

[1] "distance%=0.3267"

Ashkenazi_Jew:average

Mycenaean,28
Minoan_Lasithi,20.6
Iran_ChL,19
Polish,10.8
Levant_N,9.8
Minoan_Odigitria,6.4
French_East,5.4

Still, Bronze Age Levantine (N_Levant + CHL_Iran) is 30%, Ancient Aegean is 55%, and North/East European is around 15%.
The fit is still pretty good - 0.3267%.

Please explain why is it just over-fit to you? And why unlikely?

Also, consider all the quotes I gave, and that Jewish proselytism was very active at the time.

As for the source of Jewish demographics:


The 13th-century author Bar Hebraeus gave a figure of 6,944,000 Jews in the Roman world Salo Wittmayer Baron considered the figure convincing.[22] The figure of seven million within and one million outside the Roman world in the mid-first century became widely accepted, including by Louis Feldman. However, contemporary scholars now accept that Bar Hebraeus based his figure on a census of total Roman citizens. The figure of 6,944,000 being recorded in Eusebius' Chronicon, although this number is seen by modern scholarship as huge exaggeration.[23]: 90, 94, 104–05[24] Louis Feldman, previously an active supporter of the figure, now states that he and Baron were mistaken.[25]: 185 Philo gives a figure of one million Jews living in Egypt. John R. Bartlett rejects Baron's figures entirely, arguing that we have no clue as to the size of the Jewish demographic in the ancient world.[23]: 97–103 The Romans did not distinguish between Jews inside and outside of the Land of Israel/Judaea. They collected an annual temple tax from Jews both in and outside of Israel. The revolts in and suppression of diaspora communities in Egypt, Libya and Crete in 115–117 CE had a severe impact on the Jewish diaspora.
Source: Jewish diaspora article on wikipedia.

Erikl86
06-06-2018, 04:32 AM
Going back to nMonte2, and dropping most of the East French and Minoans, and the Polish:

[1] "distance%=0.0819 / distance=0.000819"

Ashkenazi_Jew:average

Minoan_Lasithi:I0070 34.30
Iran_ChL:I1661 20.10
French_East:French24434 19.60
Mycenaean:I9041 17.25
Levant_N:I1704 8.75

Again, same statistics:
BA Levantine - ~29%
BA Aegean people - ~52%
East French (Rhineland) - ~19%

Distance of 0.081%.

Also, the closeness between Italkim (Italian Jews) rite and Romanoites (ancient origins, Greek-speaking Jews) makes me believe perhaps Italkim has their origins in Romanoites that dwelled in the Italian peninsula from Roman times.

kingjohn
06-06-2018, 05:59 AM
the sefharadi jewish run of yours is logic :beerchug:
also i also saw it in eurogenes k13 4 populations
that balkan sefharadi need the french basque for the complete fit
you are correct that this is the european source in sefhardi {my self score 6% southwest europe in dna land and gencove
and my mom even higher} that is pointing that you are correct .....

Erikl86
06-06-2018, 06:41 AM
i also think the southern european admixture that western jews have is more
italian than greek :)
i saw some living dna results of full aschenazi jews they dont score aegean in the complete mode
but they do score south italy , tuscany, and some north italy
about the non -southern european ancestery in ashkenazim on this i agree with you more than likely french.....

When I do change to the more "traditional" admixture, of Bronze Age Levantines, using modern Tuscans as a model for the Mediterranean admixture, I get higher distance (again, nMonte2):

[1] "distance%=0.1844 / distance=0.001844"

Ashkenazi_Jew:average

Levant_N:I1704 27.20
Italian_Tuscan:NA20509 23.20
Iran_ChL:I1661 18.10
French_East:French24434 16.20
Levant_N:I1699 8.05
Italian_Tuscan:NA20799 7.25

Bronze Age Levantines: 53%
Tuscans: 30.4%
East French (Rhineland): 16.2%.

However, I have a problem with the attitude of using modern Italians as a proxy, for several reasons:
1. Tuscans might be mixed with assimilated Greek people from Italian south.
2. Modern Italians, even Tuscans, have Middle Eastern admixture from assimilated southern Italians, which might make them closer than they are.
3. If I try to model Sephardic Jews with Tuscans, they get much further away. The genetic similarity between Sephardic and Ashkenazic Jews is well documented, they cluster almost the same. So, if the Mediterranean admixture doesn't fit, it means they have different origins.

Check out Sephardic Jews with Tuscan and South French admixture (nMonte2):

[1] "distance%=0.134 / distance=0.0134"

Sephardic_Jew:average

Levant_N,35.4
Italian_Tuscan,24.2
Iran_ChL,23.2
French_South,17.2

This gives Southern French way too much admixture, and the distance is further away from say:

[1] "distance%=0.071 / distance=0.0071"

Sephardic_Jew:average

Mycenaean,34.2
Iran_ChL,20.4
Minoan_Lasithi,17.4
French_South,10.2
Minoan_Odigitria,9
Levant_N,8.8

While using ancient Aegean as the core Mediterranean component, you get the same basic Levantine admixture (29.4%), but European admixture is more shifted to Mediterranean (Aegean - 61%) vs Northern (10% French).

And the distance is now 0.07%.

Modern south Italians and Sicilians should cluster so close to Jews precisely because of my explanation:

Levantine admixture in Italians is probably due to Phoenician colonies in Sicily and South Italy dating some 3500 years ago, and then of course Carthage, which was a Phoenician empire centered in North Africa, controlled Sicily for centuries all the way to 150 BC, the last Punic War.

And later on, there was the Emirate of Sicily from 830-1090, about 260 years (almost 10 generations of Arab-Berber rule).

I would say a lot of Levantine, North African (J2 and E1b1 haplogroups) admixture in Sicily and southern Italy, along side Greek populations. Hence why modern Sicilians would cluster close to Jews. Add to that the fact that until 1800s they were ruled by a Norman kingdom from Northern Europe, which resulted in about 7% of Sicilians carrying North European haplogroups.

Mycenaean Greeks + Near Easterns + small amount of North Europeans, would indeed make them cluster even closer to Ashkenazi Jews than modern Greek Islanders would, and this is what we see.

eolien
06-06-2018, 06:58 AM
several questions/comments:

1- I've also read somewhere that a Roman census gave 1M Jews but please take into account that the census concerns only Roman citizens.
2- Where did you get your sefardi samples from?
3- I cannot really comment on your statistical results BUT don't you need some kind of positive and negative controls to make your case stronger? I personally would take some Levantine populations and see how they deviate (for example Cyprus, Malta, Christian Lebanese, Druze)

thank you.

kingjohn
06-06-2018, 07:01 AM
nice :) so you still need some southern french 10% for the sefhardi no matter
what you use in nmonte2 as the distance is shorter
that is cool
i copy paste it those are nice runs by you B)
regards
Adam

p.s
personally for me it is relevant because i am aschenazi/sefhardi mix B)

Erikl86
06-06-2018, 08:00 AM
several questions/comments:

1- I've also read somewhere that a Roman census gave 1M Jews but please take into account that the census concerns only Roman citizens.
2- Where did you get your sefardi samples from?
3- I cannot really comment on your statistical results BUT don't you need some kind of positive and negative controls to make your case stronger? I personally would take some Levantine populations and see how they deviate (for example Cyprus, Malta, Christian Lebanese, Druze)

thank you.

1 - regarding the census - even if they were not 25% or even 10%, there are endless accounts, both Jewish, and non-Jewish, about huge amount of mainly Greek converts to Hellenistic Judaism.
2 - I got it from the Eurogenes blog itself.
3 - I actually have tried different Levantine admixtures instead of the classical Bronze Age model (although I think it is the most reliable). Check it out:

First of all, I've tried just plain 'ol simple Neolithic Levant + Tuscan + N. European:

[1] "distance%=12.6948 / distance=0.126948"

Ashkenazi_Eastern_Euro

N_Levant_Avg 50.6
IT_Tuscany 35.0
Polish_average 14.3

Got a huge distance of 12.8%.

Then replaced the mixture of N_Levant / CHL_Iran with Samaritans, the closest living people to ancient Canaanites, and got wacko results:

[1] "distance%=0.4041 / distance=0.004041"

Ashkenazi_Jew:average

Samaritan:GSM537032 61.5
French_East:French24434 38.5

Obviously Ashkenazi Jews aren't 60% Samaritans, 40% Rhineland French.

And that result didn't change regardless if I put Tuscans or Aegean Mediterranean admixture.

So, I've added some Caucasian admixture which the Samaritans seem to lack (but should exist, according to my theory, in Jews from early on because of admixture with Philistines, which would have Steppes and Neolithic Anatolian admixture if they are indeed Mycenaean-influenced descendants of Minoans from Crete). I've tried more ancient ones:

[1] "distance%=0.2488 / distance=0.002488"

Ashkenazi_Jew:average

Samaritan:GSM537032 32.4
Italian_Tuscan:NA20509 17.8
Italian_Tuscan:NA20799 17.6
Anatolia_BA:I2499 13.0
Armenia_EBA:I1635 9.7
French_East:French24434 9.6

Surprise surprise - Samaritans got about ~32% - similar to what I got when using Aegean people, and the result is much closer. But, I have two problems: Rhineland French is now way too low for Ashkenazi, about 9.6%. I blame Tuscans, because they might have higher level of Germanic admixture and perhaps direct Bronze Age Anatolian admixture communicated via some assimilated Greek people from south Italy.

Then I've tried to model Mycenaeans, to see if perhaps they were mixed with Levantines which is why they would show up as 50-60% of Western Jews:

[1] "distance%=0.2614"

Mycenaean:average

Minoan_Lasithi,59.2
Minoan_Odigitria,15.8
ArmeniaChL,7.6
Srubnaya_Spiridonovka,5.2
AnatoliaChL,4.8
Srubnaya_Rozhdestveno,3.2
Srubnaya_Barinovka,2.2
Srubnaya_Novoselki,1.2
Srubnaya_Uvarovka,0.8


Srubnaya is Bronze Age Steppes culture which dates some 4000 years ago, around the time I've read the Stepped people immigrated to the Balkans and mixed with Minoans, forming Mycenaean people.

CHL Armenians and CHL Anatolian represent earlier immigration to the Aegean and south Balkan areas.

Then, If I add any pre-Bronze Age Levantine admixture, for example Neolithic Levant or the admixture of CHL Iranians and Neolithic Levantines (which would basically mean Bronze Age Levantines), the distance grows dramitcally, check it out:

[1] "distance%=2.7243"

Mycenaean:average

Minoan_Lasithi,55.2
Minoan_Odigitria,17.8
Srubnaya_Rozhdestveno,4.4
Srubnaya_Barinovka,4
Levant_N,3.6
AnatoliaChL,3.4
Srubnaya_Spiridonovka,3.4
ArmeniaChL,2.8
Iran_ChL,2.2
Srubnaya_Novoselki,1.8
Srubnaya_Uvarovka,1.4

Adding Bronze Age Levantines (N_Levant + CHL_Iran), the distance grows about x10 times, and even then those are pushed down to mere 5.8%.

This means that current Levantine admixture found in modern Greek Islanders is post-Neolithic, post-Mycenaean and was most likely the result of Hellenistic West-East mixture (which also created Western Jews).

Then I've found an article which actually corroborates this small Levantine admixture, coming from Neolithic Anatolians to Minoans and Mycenaeans:


"The Minoans could be modelled as a mixture of the Anatolia
Neolithic-related substratum with additional ‘eastern’ ancestry, but
the other two groups had additional ancestry: the Mycenaeans had
approximately 4–16% ancestry from a ‘northern’ ultimate source related
to the hunter–gatherers of eastern Europe and Siberia (Table 1), while
the Bronze Age southwestern Anatolians may have had ~ 6% ancestry
related to Neolithic Levantine populations. "


So, less than 6%, through Anatolians. (I can't add links, so the name of the article is "Genetic origins of the Minoans and Mycenaeans".).

eolien
06-06-2018, 09:00 AM
double post

eolien
06-06-2018, 09:03 AM
Perhaps you misunderstood me. I was thinking why not run similar tests on Cyprus, Malta, Christian Lebanese, Druze INSTEAD of Ashkenazi and Sefaradi. I am also of the opinion that the reference samples chosen for these populations could already be selected to give a coherent picture. I for example fall exactly in the middle of Sefardi cluster in G25 PCA but it can be that many Ashkenazi samples would not show such a tight picture. Moreover, you still need to explain why we are close to South italians and not to Greeks et al.

23703

Erikl86
06-06-2018, 09:12 AM
Perhaps you misunderstood me. I was thinking why not run similar tests on Cyprus, Malta, Christian Lebanese, Druze INSTEAD of Ashkenazi and Sefaradi. I am also of the opinion that the reference samples chosen for these populations could already be selected to give a coherent picture. I for example fall exactly in the middle of Sefardi cluster in G25 PCA but it can be that many Ashkenazi samples would show such a tight picture. Moreover, you still need to explain why we are close to South italians and not to Greeks et al.

23703

Oh I see. I'll run some tests then.

As for the explanation why South Italians (and Sicilians especially) are so close to Western Jews, and especially Ashkenazi Jews, rather than Greek Islanders (mainland Greeks are mixed with Slavic people since early Medieval Ages and so Greek Islanders are closer to south Italians than to mainland Greeks in many occasions).

About 37% of Sicilian background comes from Greek colonists which heavily populated southern Italy and Sicily since the 8th century BC. Greek was the majority language in southern Italy and Sicily until Medieval Ages. Greek people weren't expelled, but were Latinized and absorbed by the Latin-speaking people in Italy after the Byzantine Empire lost it's territories in the Italian peninsula.



Modern south Italians and Sicilians should cluster so close to Jews precisely because of my explanation:

Levantine admixture in Italians is probably due to Phoenician colonies in Sicily and South Italy dating some 3500 years ago, and then of course Carthage, which was a Phoenician empire centered in North Africa, controlled Sicily for centuries all the way to 150 BC, the last Punic War.

And later on, there was the Emirate of Sicily from 830-1090, about 260 years (almost 10 generations of Arab-Berber rule).

I would say a lot of Levantine, North African (J2 and E1b1 haplogroups) admixture in Sicily and southern Italy, along side Greek populations. Hence why modern Sicilians would cluster close to Jews. Add to that the fact that until 1800s they were ruled by a Norman kingdom from Northern Europe, which resulted in about 7% of Sicilians carrying North European haplogroups.

Mycenaean Greeks + Near Easterns + small amount of North Europeans, would indeed make them cluster even closer to Ashkenazi Jews than modern Greek Islanders would, and this is what we see.

Erikl86
06-06-2018, 11:29 AM
I was thinking why not run similar tests on Cyprus, Malta, Christian Lebanese, Druze INSTEAD of Ashkenazi and Sefaradi.

Few thoughts regarding checking for Minoan or Mycenaean admixture in those populations though:

Christian Lebanese should have some Aegean admixture, because the region was heavily Hellenized. I would expect Greek Orthodox Christians to be more Hellenized than Maronites.

Judeans would mix with Philistines because they bordered each other. Philistines disappeared and were absorbed into local Southern shores of the Levant long after the Kingdom of Israel was already abolished by the Assyrians, and when Samaritans began to form their own separate identity, roughly around the 5-6th centuries BC.

Also regarding Samaritans - I am not aware of massive conversion to Samaritanism by Greeks, as opposed to Judaism. So if there is no admixture of Aegean with Samaritans, that would further prove their claim to be descendants of the Northern Kingdom of Israel, rather than Judeans which mixed with immigrants from Assyria as our Bible claims they are.

Cypriots would appear somewhat close to ancient Greeks but also heavily Levantine because there was a very long presence of Phoenicians on that island.

So actually, aside from perhaps Samaritans, which should have some low or nearly no Greek admixture, these are my predictions.

Now for the results:

Let's see about Christian Lebanese:

[1] "distance%=0.0292 / distance=0.000292"

Lebanese_Christian:average

Minoan_Lasithi:I0070 31.55
Iran_ChL:I1661 25.15
Levant_N:I1704 24.50
Iran_ChL:I1665 12.55
Mycenaean:I9041 4.30
Levant_N:I1699 1.65
Iran_ChL:I1662 0.30

Roughly 65% Bronze Age Levantines, 35% ancient Aegean. My bet, these are Greek Orthodox Christian Lebanese which have been sampled

Distance of 0.0292%.

Let's check Samaritans, with the exact same populations as Christian Lebanese:

[1] "distance%=0.5681 / distance=0.005681"

Samaritan:GSM537032

Iran_ChL:I1665 34.5
Levant_N:I1704 33.2
Levant_N:I0867 32.4

No Aegean admixture appeared.

Let's check Christian Israeli Arabs:

[1] "distance%=0.134 / distance=0.00134"

Arab_Israel_1:average

Levant_N:I1704 30.2
Minoan_Lasithi:I0073 26.6
Iran_ChL:I1665 21.8
Iran_ChL:I1662 15.9
Iran_ChL:I1661 5.5

Roughly 73% Bronze Age Levantines, 26% Aegean - mostly Minoan (No Mycenaean).

Next, let's check Druze:

[1] "distance%=0.0289 / distance=0.000289"

Druze:average

Minoan_Lasithi:I0073 21.20
Levant_N:I1704 20.35
Iran_ChL:I1665 19.65
Iran_ChL:I1661 14.40
Mycenaean:I9010 9.15
Iran_ChL:I1662 8.05
Mycenaean:I9006 6.50
Iran_ChL:I1674 0.70

Very similar to Christian Lebanese - Bronze Age Levantine is 63%, Ancient Aegean is 37% (both Minoan and Mycenaean). Distance is 0.0289%.

Now let's check Cypriots:

[1] "distance%=0.0998 / distance=0.000998"

Cypriot:average

Minoan_Odigitria:I9131 24.80
Mycenaean:I9041 21.40
Minoan_Lasithi:I0073 17.10
Iran_ChL:I1661 12.00
Iran_ChL:I1662 8.65
Iran_ChL:I1665 7.25
Levant_N:I1704 6.70
Iran_ChL:I1674 2.10

Mirror image of Druze :). 63% Aegean, 37% Levantines. Not surprising at all.

All in all, it seems all non-Muslim Levantines I've examined are mixed with Aegean people to some degree 25-35% roughly, except for Samaritans which yield zero admixture with them. I would say this is probably the result of Hellenistic immigration of Greeks which settled in the Levant, probably mostly from the islands because these were closer (and perhaps even before the Hellenistic period). Palestinian Christians seem to be less mixed with Aegeans than Lebanese Christians, but I suspect that the average is biased by Greek Orthodox Lebanese. Maybe if I will examine only Maronites, it would yield closer results to Palestinian Christians (or even less so).

Western Jews probably are much more Greek because we are the result of already mixed Levantine-Greek ancestry, leaving and marrying Greek women converts in the Hellenistic world.

Now, let's examine Muslims Levantines.

First, Muslim Lebanese:

[1] "distance%=0.7972 / distance=0.007972"

Lebanese_Muslim:average

Iran_ChL:I1674 38.5
Mycenaean:I9010 29.9
Levant_N:I0867 18.9
Levant_N:I1699 12.8

Lebanese Muslims are further away - 0.79% vs. 0.029% from this admixture, and seem to have lower Aegean admixture.

Now let's check Muslim Palestinian:

[1] "distance%=2.3663 / distance=0.023663"

Palestinian:average

Levant_N:I0867 45.2
Iran_ChL:I1674 35.0
Mycenaean:I9010 19.8

Wow - really far away compare to the others. It is obviously mainly Levantine, with some Aegean admixture, but these populations are not enough to get accurate model.

Now I think I know what is missing - sub-Saharan admixture which is found among the Middle East only among Arab Muslims, but not Christians or Jews etc.

This sub-Saharan admixture comes from two sources: south Arabian admixture, and East African female sex slaves.

Let's see, if I add Bedouin to the admixture, if that helps:

distance%=1.5895 / distance=0.015895"

Palestinian:average

BedouinB:average 46.3
Mycenaean:I9010 29.4
Iran_ChL:I1674 24.2

Bedouin improves it, but not a lot.

Unfortunately I don't have sub-Saharan African samples with Basal-rich K7 stats, and I've tried any other combination - Saudis which should have higher sub-Saharan African admixture, even Christian Palestinians. None of those yield less than 1.3%. If I've taken out the Aegean, it would jump to +3%, so it seems to be partially Aegean as well.

Let's try to add Saudi and Bedouin to the Muslim Lebanese to see if it'll help, because Lebanese Muslims are the least admixed with recent sub-Saharan African genes:

[1] "distance%=0.0979 / distance=0.000979"

Lebanese_Muslim:average

Mycenaean:I9010 29.8
Iran_ChL:I1674 29.2
BedouinB:HGDP00607 28.4
Mycenaean:I9006 10.8
Saudi:SaudiA9 1.9

I guess Bedouins, depends on where they were collected, is already mixed with Levantines and Saudis (and probably takes all the Neolithic Levantine component). Obvious south Middle Easterners like Saudi is roughly 2%. This makes sense. See that Levantine and Arabian is 60%, Aegean is 40%. Distance is 0.097%. Basically, they are Lebanese Christians mixed with Arabians, as expected.

Agamemnon
06-06-2018, 11:40 AM
I think this exercise is rather futile, using the exact same method one could easily argue that the present-day Lebanese owe the vast majority of their ancestry to an Anatolia_BA source for example (this is more likely to be the case in Cyprus however). In the same vein, it's very likely the Philistines will turn out to be very similar to the Mycenaean samples we have, if that's the case one could also make the argument that Western Jews owe most of their ancestry to the Philistines. On closer inspection, such models are undoubtedly bound to fall apart, which is exactly what I expect to see once we get more ancient data from the Mediterranean.

Also in many of the fits you've obtained for Levantine populations, replacing Iran_Chl and Levant_N with Levant_BA would be more accurate when populations such as the Mycenaeans and the Minoans are thrown in (for Palestinians though, additional African admixture is absolutely necessary to obtain convincing fits).

Edit: I can understand the appeal of such models, after all these ancient Eastern Mediterranean populations are extremely close to Western Jews (I was able to model my father as ~90% Mycenaean for example) and could potentially provide a straightforward explanation for the ethnogenesis of Western Jewry (which was Greek-speaking after all). We are literally grasping at straws with the limited data however, this is just the beginning and I have no doubt that future ancient data from the Mediterranean will make things a lot clearer.

talljimmy0
06-06-2018, 12:01 PM
The answer lies in Alexandria. The City of London of the ancient world.

eolien
06-06-2018, 12:02 PM
I am really confused. Why are Iran_ChL is considered part of Levant? Why Sidon samples are not used, why some populations prefer Minoan over Mycenaean?

And finally all the conclusions by Erikl86 are already biased because he tries to interpret/justifiy the results by assumed historical events. This is not a valid scientific method. For example Samaritans were known to be a large population in the common era. Therefore we expect many Palestinian Muslims to have origins in Samaritans. The same for Bedouins. Likewise Maronites and Druze are religious groups and not ethnic groups, but whatever their origins they are assumed to remain relatively isolated since the middle ages.

Erikl86
06-06-2018, 12:06 PM
I think this exercise is rather futile, using the exact same method one could easily argue that the present-day Lebanese owe the vast majority of their ancestry to an Anatolia_BA source for example (this is more likely to be the case in Cyprus however). In the same vein, it's very likely the Philistines will turn out to be very similar to the Mycenaean samples we have, if that's the case one could also make the argument that Western Jews owe most of their ancestry to the Philistines. On closer inspection, such models are undoubtedly bound to fall apart, which is exactly what I expect to see once we get more ancient data from the Mediterranean.

Also in many of the fits you've obtained for Levantine populations, replacing Iran_Chl and Levant_N with Levant_BA would be more accurate when populations such as the Mycenaeans and the Minoans are thrown in (for Palestinians though, additional African admixture is absolutely necessary to obtain convincing fits).

Edit: I can understand the appeal of such models, after all these ancient Eastern Mediterranean populations are extremely close to Western Jews (I was able to model my father as ~90% Mycenaean for example) and could potentially provide a straightforward explanation for the ethnogenesis of Western Jewry (which was Greek-speaking after all). We are literally grasping at straws with the limited data however, this is just the beginning and I have no doubt that future ancient data from the Mediterranean will make things a lot clearer.

First, I absolutely agree we need more Mediterranean ancient data.

However, I've seen people using modern populations as admixture to try and find the right fit for Western Jews.

For me, this model actually fits historical logic, not only pure numbers.

I would be very surprised if there was no Mycenean or Minoan admixture among other Levantines.

I think that using modern populations, other than Northern Europeans for Ashkenazi (because that admixture got added in the last 750-1000 years, and populations in Western Europe did not change a lot since then), mixed with ancient ones, to try to show how the Mediterranean admixture is actually Italian, also requiring some exotic Caucasus admixture, is inaccurate.

These admixtures would happen +2000 years ago. This is my case - that Western Jews were already admixed when they left the Levant after 70 AD.

So there is no point in saying Samaritan + Tuscan + Caucasian population = Jewish, or replacing Samaritan with Druze or Christian Lebanese.

I've already mentioned my personal problems with using modern Mediterranean populations in such models - Tuscans today might resemble Romans from +2000 years ago, but they might not.

Mycenaean and Minoans, lacking significant Neolithic Levantine admixture, provide better sample - and they seem to yield good results as well.

Also, I have thought about using Bronze Age Levantines, instead of the model of Neolithic Levantine and CHL Iranian, but there's a problem with it: Bronze Age in the Levant ended around the 13th century BC. There were already Greek colonies and Phoenician (Bronze Age Levantine) colonies spread throughout Eastern Mediterranean, and there were already Bronze Age Anatolians infiltrating the Levant (Hurrians, Mittani etc.). So to make sure they don't overlap, I adopted this model (which I didn't invent, but took from the article "Continuity and Admixture in the Last Five Millennia of Levantine History from Ancient Canaanite and Present-Day Lebanese Genome Sequences").

Erikl86
06-06-2018, 12:23 PM
And finally all the conclusions by Erikl86 are already biased because he tries to interpret/justifiy the results by assumed historical events.

Erm, nope. I try to assume what would be the result, based on known historical events. Using Tuscans to model ancient Mediterraneans and claim they must have mixed with Jews to form Western Jews, thus ignoring mass conversions of Hellenistic and Roman Greeks to Judaism, is much worse science. Also, ignoring the fact that southern Italians and Sicilians are heavily Greek and mixed with Levantines, thus perhaps cluster close to Western Jews rather than the fact that they resemble Romans, is also bad science.


I am really confused. Why are Iran_ChL is considered part of Levant? Why Sidon samples are not used

I've explained I don't use the Sidon (which is bronze age Levantine) for the following reasons:


I have thought about using Bronze Age Levantines, instead of the model of Neolithic Levantine and CHL Iranian, but there's a problem with it: Bronze Age in the Levant ended around the 13th century BC. There were already Greek colonies and Phoenician (Bronze Age Levantine) colonies spread throughout Eastern Mediterranean, and there were already Bronze Age Anatolians infiltrating the Levant (Hurrians, Mittani etc.). So to make sure they don't overlap, I adopted this model (which I didn't invent, but took from the article "Continuity and Admixture in the Last Five Millennia of Levantine History from Ancient Canaanite and Present-Day Lebanese Genome Sequences").

Also, the reason for using Neolithic Levantine and CHL Iranian is from that Hebar et al article, "Continuity and Admixture in the Last Five Millennia of Levantine History from Ancient Canaanite and Present-Day Lebanese Genome Sequences":


We find that a Bronze Age Canaanite-related ancestry was widespread in the region, shared among urban populations inhabiting the coast (Sidon) and inland populations (Jordan) who likely lived in farming societies or were pastoral nomads. This Canaanite-related ancestry derived from mixture between local Neolithic populations and eastern migrants genetically related to Chalcolithic Iranians. We estimate, using linkage-disequilibrium decay patterns, that admixture occurred 6,600–3,550 years ago, coinciding with recorded massive population movements in Mesopotamia during the mid-Holocene.

And also:


Lazaridis et al.13 reported that Jordan_BA can be modeled as mixture of Neolithic Levant (Levant_N) and Chalcolithic Iran (Iran_ChL). We computed the statistic f4(Levant_N, Sidon_BA; Ancient Eurasian, Chimpanzee) and found that populations from the Caucasus and ancient Iran shared more alleles with Sidon_BA than with Neolithic Levant (Figure 2A and S10). We then used qpAdm8 (with parameter allsnps: YES) to test whether Sidon_BA can be modeled as mixture of Levant_N and any other ancient population in the dataset and found good support for the model of Sidon_BA being a mixture of Levant_N (48.4% ± 4.2%) and Iran_ChL (51.6% ± 4.2%) (Figure 2B; Table S3).

As for your other question:

why some populations prefer Minoan over Mycenaean?

I think that it's because Minoans and Mycenaeans are very close genetically, with Mycenaeans basically being 75% Minoans + Eurasian Steppes people. If to quote Lazaridis himself, the author of the study of the remains of those Mycenaeans and Minoans:


"The study results show that Minoans and Mycenaeans were genetically highly similar — but not identical — and that modern Greeks descend from these populations.

The Minoans and Mycenaeans descended mainly from early Neolithic farmers, likely migrating thousands of years prior to the Bronze Age from Anatolia, in what is today modern Turkey.

“Minoans, Mycenaeans, and modern Greeks also had some ancestry related to the ancient people of the Caucasus, Armenia, and Iran,” said co-lead author Dr. Iosif Lazaridis, a postdoctoral researcher at Harvard Medical School."

So perhaps nMonte is having difficulties distinguishing between the two.
Another possibility, is that most admixture occurred between Aegean Islanders, rather than immigrants from mainland Greece, because they were closer to the Levant geographically. These people would be basically Minoans which have been "Mycenaean" linguistically and culturally, such as the people of Crete.
Another possibility, is that some of those populations above, especially Christian Palestinians, are descendants from Judeans, which according to my theory, their first admixture with Aegean people would be Philistines, which according to most popular theories, originated from Crete and thus would be genetically Minoan.

Agamemnon
06-06-2018, 12:26 PM
Please explain why is it just over-fit to you? And why unlikely?

For largely the same reason a fit such as this should be considered overfitted (if not downright inaccurate)... G25 w/ pen=0:

[1] "distance%=1.554"

Lebanese_Druze

Armenia_EBA,37.2
BedouinB,26.8
Mycenaean,21.2
Sidon_BA,10.8
Minoan_Lasithi,4


Also, consider all the quotes I gave, and that Jewish proselytism was very active at the time.

Jewish proselytism was restricted to very specific episodes, such as the expansion of the Hasmonean kingdom (which brought about the forced conversions of the Idumeans, the Itureans and many Samaritans), the conversion of the Himyarites to Judaism or the conversion of the Khazars to Judaism. Such episodes were the exception, not the rule. As a rule of thumb, Judaism was (and still is) much more akin to the religions of the ancient world since the entry was primarily biological.


First, I absolutely agree we need more Mediterranean ancient data.

However, I've seen people using modern populations as admixture to try and find the right fit for Western Jews.

For me, this model actually fits historical logic, not only pure numbers.

I would be very surprised if there was no Mycenean or Minoan admixture among other Levantines.

I think that using modern populations, other than Northern Europeans for Ashkenazi (because that admixture got added in the last 750-1000 years, and populations in Western Europe did not change a lot since then), mixed with ancient ones, to try to show how the Mediterranean admixture is actually Italian, also requiring some exotic Caucasus admixture, is inaccurate.

These admixtures would happen +2000 years ago. This is my case - that Western Jews were already admixed when they left the Levant after 70 AD.

So there is no point in saying Samaritan + Tuscan + Caucasian population = Jewish, or replacing Samaritan with Druze or Christian Lebanese.

I've already mentioned my personal problems with using modern Mediterranean populations in such models - Tuscans today might resemble Romans from +2000 years ago, but they might not.

Mycenaean and Minoans, lacking significant Neolithic Levantine admixture, provide better sample - and they seem to yield good results as well.

Also, I have thought about using Bronze Age Levantines, instead of the model of Neolithic Levantine and CHL Iranian, but there's a problem with it: Bronze Age in the Levant ended around the 13th century BC. There were already Greek colonies and Phoenician (Bronze Age Levantine) colonies spread throughout Eastern Mediterranean, and there were already Bronze Age Anatolians infiltrating the Levant (Hurrians, Mittani etc.). So to make sure they don't overlap, I adopted this model (which I didn't invent, but took from the article "Continuity and Admixture in the Last Five Millennia of Levantine History from Ancient Canaanite and Present-Day Lebanese Genome Sequences").

I think you might be overestimating the impact the Greeks had in the Levant, I mean outside the Philistine coast (which was heavily colonised by Aegean settlers) and some cities like Caesarea I very much doubt we're dealing with a massive phenomenon. The fact of the matter is that the most parsimonious model remains a mixture of Levantines and North-Central Italians with additional local admixture. I've also managed to replicate Haber et al.'s model (Levant_N + Iran_Chl) to proxy the Levantine ancestry of Western Jews using the very same mixture.

Also, there is no reason to assume that the Philistines were genetically Minoan, if anything they'll probably be somewhere between Mycenaeans and Anatolia_EBA.

Erikl86
06-06-2018, 12:34 PM
For largely the same reason a fit such as this should be considered overfitted (if not downright inaccurate)... G25 w/ pen=0:

[1] "distance%=1.554"

Lebanese_Druze

Armenia_EBA,37.2
BedouinB,26.8
Mycenaean,21.2
Sidon_BA,10.8
Minoan_Lasithi,4




And what about these:


[1] "distance%=0.3129 / distance=0.003129"

Ashkenazi_Jew:average

Samaritan 52.70
Italian_Bergamo:average 31.55
Lithuanian:average 8.35
Avar:average 7.40


[1] "distance%=0.3732 / distance=0.003732"

Ashkenazi_Jew:average

Druze:average 61
Italian_Bergamo:average 39

Are they more accurate? Are they less over-fitted? Why would you have modern Avar admixture in Western Jews? Can you model Sephardic Jews in similar ways (just without the Lithuanian)? And 60/40 for Samaritans/Bergamo-Italians is way, way inaccurate, plus Bergamo Italians would have a lot of Celtic and even some Germanic admixture - they are not a good example of Mediterranean component.

I actually thank the member which requested me to check the other populations (Druze, Samaritans etc.), because it shows consistence. I would be very surprised, if after +3000 years of Aegean interactions with the Levant, from ancient Greece to Hellenistic times all the way to Eastern Roman Empire, there would be little Aegean admixture.

I'm all for refuting these findings - but just claiming they are inaccurate, without explaining why, or claim they have no historical proof - tell me, why. My basic assumption is that the main admixture of Western Jews with Europeans, happened before they even left the Levant, and before the Roman abolition of the 2nd Temple.


Jewish proselytism was restricted to very specific episodes, such as the expansion of the Hasmonean kingdom (which brought about the forced conversions of the Idumeans, the Itureans and many Samaritans), the conversion of the Himyarites to Judaism or the conversion of the Khazars to Judaism. Such episodes were the exception, not the rule. As a rule of thumb, Judaism was (and still is) much more akin to the religions of the ancient world since the entry was primarily biological.

Agreed, but Judaism and semi-Judaism was very popular among Greeks in late Hellenistic times and 1st centuries. This is very well documented, including in the Talmud. I've brought numerous sources about it. It's actually more documented than either Himyar or Khazaria.

Agamemnon
06-06-2018, 12:50 PM
And what about these:


Are they more accurate? Are they less over-fitted? Why would you have modern Avar admixture in Western Jews? Can you model Sephardic Jews in similar ways (just without the Lithuanian)? And 60/40 for Samaritans/Bergamo-Italians is way, way inaccurate, plus Bergamo Italians would have a lot of Celtic and even some Germanic admixture - they are not a good example of Mediterranean component.

They're actually far more accurate, just like this fit (which is the same as the one I posted above, just without pen=0) is more accurate:

[1] "distance%=2.1309"

Lebanese_Druze

Sidon_BA,65.8
Armenia_EBA,16.4
Anatolia_BA,9.8
BedouinB,7.2
Mycenaean,0.8


^^Notice how the distance is greater than in the first fit.

As for the Avar admixture, it is in many cases necessary. Samaritans closely resemble BA_Canaanites, in comparison modern-day Levantines are shifted towards the Caucasus, this is why the inclusion of the Avars enables us to obtain a better fit. Another way of doing this is by adding Armenia_EBA, just look at the fit above. What we will probably see in upcoming results is that the post-LBA collapse Levant saw a slight shift in the direction of the Caucasus and Mesopotamia. The inclusion of the Druze instead of the Samaritans was meant to demonstrate both this shift and the fact that Eastern European admixture might vary according to the Levantine proxy being used. You will also notice that I am only using present-day populations in these fits, as opposed to mixing ancient and modern ones. And yes, you can actually proxy other Western Jewish populations using the exact same model.


I actually thank the member which requested me to check the other populations (Druze, Samaritans etc.), because it shows consistence. I would be very surprised, if after +3000 years of Aegean interactions with the Levant, from ancient Greece to Hellenistic times all the way to Eastern Roman Empire, there would be little Aegean admixture.


Many of the fits you obtained are extremely overfitted, that's the only consistence to be found actually.


Agreed, but Judaism and semi-Judaism was very popular among Greeks in late Hellenistic times and 1st centuries. This is very well documented, including in the Talmud. I've brought numerous sources about it. It's actually more documented than either Himyar or Khazaria.

"Semi-Judaism" is a rather ambiguous term, for example Josephus described the Samaritans as "half-Jews".

Erikl86
06-06-2018, 01:37 PM
They're actually far more accurate, just like this fit (which is the same as the one I posted above, just without pen=0) is more accurate:

[1] "distance%=2.1309"

Lebanese_Druze

Sidon_BA,65.8
Armenia_EBA,16.4
Anatolia_BA,9.8
BedouinB,7.2
Mycenaean,0.8


^^Notice how the distance is greater than in the first fit.

As for the Avar admixture, it is in many cases necessary. Samaritans closely resemble BA_Canaanites, in comparison modern-day Levantines are shifted towards the Caucasus, this is why the inclusion of the Avars enables us to obtain a better fit. Another way of doing this is by adding Armenia_EBA, just look at the fit above. What we will probably see in upcoming results is that the post-LBA collapse Levant saw a slight shift in the direction of the Caucasus and Mesopotamia. The inclusion of the Druze instead of the Samaritans was meant to demonstrate both this shift and the fact that Eastern European admixture might vary according to the Levantine proxy being used. You will also notice that I am only using present-day populations in these fits, as opposed to mixing ancient and modern ones. And yes, you can actually proxy other Western Jewish populations using the exact same model.



I strongly disagree regarding the model for Druze. First, Bedouins are a modern population. Second - why would you add Mycenaean ? on what grounds? Also, Early Bronze Age Armenians and Bronze Age Anatolians highly resemble Aegean and Minoans. Also 25% Bronze Age Caucasians and Anatolians is too high, IMO, and Mycenaean and Minoan admixture gives you much better results.

I wonder, how do people explain admixture of Levantines among modern Greek Islanders, while tend to reject the notion of Greek admixture with Levantines, when in fact Greek colonization in the Eastern Mediterranean was very prominent from at least the 10th century BC, onward.

As for the Avars - do you know they are mainly Caucasian and haven't mixed with Turks or Iranians or Eurasian Steppes people in the last +2500 years? Why not use Dagestani? Or Circassians? Because it just gives you a better %% ? Also, this model might be more seasonable than the ones I've quoted which you provided more than a year ago - Samaritans & Bergamo Italians? and 60 / 40 ? Bergamo are highly Celtic. And the result you got 0.31% - is, according to the standards you provide, a very good example of over-fitting. To me, just throwing a bunch of different ethnicities without any historical background, or very far-fetched historical background, is just inaccurate.

The only modern ones I care to use, is the Northern European admixture, for a very good reason - according to the study I've quoted in the original post, the admixture with Northern Europeans happened just after the population bottleneck Ashkenazi Jews suffered from - in the last 20 generations. After the 10th century, there were very little mass movements of whole ethnic groups into Western Europe. So basically, German from 750 years ago would resemble genetically a German today. Modern Avars would very likely not resemble BA Armenians and Anatolians.




Many of the fits you obtained are extremely overfitted, that's the only consistence to be found actually.


Nope, they're also pretty consistent. Aegean and Levantine admixture is consistent in all of them. If I would see in Druze 60% Aegean but Lebanese Christians would have 20%, I would say there is some inconsistency. This is not the case. And Samaritans, according to my model, are indeed pretty much identical to Bronze Age Canaanites (and seem to lack any Aegean admixture).



"Semi-Judaism" is a rather ambiguous term, for example Josephus described the Samaritans as "half-Jews".

Samaritans are not the half Jews. I'm talking about Greek "God fearers" and Greek converts which joined Hellenistic Jewish communities all around the East Mediterranean. Or do you deny their existence?

Erikl86
06-06-2018, 02:07 PM
I think you might be overestimating the impact the Greeks had in the Levant, I mean outside the Philistine coast (which was heavily colonised by Aegean settlers) and some cities like Caesarea I very much doubt we're dealing with a massive phenomenon.

Actually, I don't know how they couldn't have a demographic impact on the Levant. They have been colonizing it since at least the 10th century BC, and there was even a Minoan settlement found in Northern Israel:

The Minoans were traders, and their cultural contacts reached Egypt's Old Kingdom, copper-containing Cyprus, Canaan and the Levantine coast and Anatolia. In late 2009 Minoan-style frescoes and other artifacts were discovered during excavations of the Canaanite palace at Tel Kabri, Israel, leading archaeologists to conclude that the Minoan influence was the strongest on the Canaanite city-state.
Source: wikipedia's article on Minoan civilization.

And also second wave of Greek colonization:


More than thirty Greek city-states had multiple colonies around the Mediterranean world, with the most active being Miletus, of the Ionian League, with ninety colonies stretching throughout the Mediterranean Sea, from the shores of the Black Sea and Anatolia (modern Turkey) in the east, to the southern coast of the Iberian Peninsula in the west, as well as several colonies on the Libyan coast of northern Africa,[11] from the late 9th to the 5th centuries BC.

There were two similar types of colony, one known as an ἀποικία - apoikia (pl.: ἀποικίαι, apoikiai) and the other as an ἐμπόριov - emporion (pl.: ἐμπόρια, emporia). The first type of colony was a city-state on its own; the second was a Greek trading-colony.

The Greek city-states began establishing colonies around 900[12] - 800 BC, at first at Al Mina on the coast of Syria and the Greek emporium Pithekoussai at Ischia in the Bay of Naples, both established about 800 BC by Euboeans.[13]

23709

While indeed, before the 4th century BC, most Greek colonization in East Mediterranean was not in specifically in the Levant (but also existed there), after the 4th century BC, and all the way to the 8th century AD, for more than 1200 years, there have been Greek people coming to the Levant.

After 300 BC, the Hellenistic colonization began. Dark Blue - Greek-majority areas, Pale Blue - highly Hellenized local population, around 300 BC:

23708

And this is what wikipedia has to say:


The history of Greeks in Syria traditionally begins with Alexander the Great's conquest of the Persian Empire. In the aftermath of Alexander's death, his empire was divided into several successor states, and thus ushering in the beginning of the Hellenistic Age. For the Levant and Mesopotamia, it meant coming under the control of Seleucus I Nicator and the Seleucid Empire. The Hellenistic period was characterized by a new wave of Greek colonization.[5] Ethnic Greek colonists came from all parts of the Greek world, not, as before, from a specific "mother city".[6] The main centers of this new cultural expansion of Hellenism in the Levant were cities like Antioch, and the other cities of the Tetrapolis Seleukis. The mixture of Greek-speakers gave birth to a common Attic-based dialect, known as Koine Greek, which became the lingua franca throughout the Hellenistic world.

And:


The Seleucid Empire was a major empire of Hellenistic culture that maintained the pre-eminence of Greek customs in which a Greek political elite dominated, in newly founded urban areas.[7][8][9][10] The Greek population of the cities who formed the dominant elite were reinforced by emigration from Greece.[7][8] The creation of new Greek cities were aided by the fact that the Greek mainland was overpopulated and therefore made the vast Seleucid Empire ripe for colonization. Apart from these cities, there were also a large number of Seleucid garrisons (choria), military colonies (katoikiai) and Greek villages (komai) which the Seleucids planted throughout the empire to cement their rule.


Come to think about it, it actually makes a lot of sense why Samaritans lack any significant Greek admixture. By the 4th century BC, they were already a closed group of people, separated from Jews. Any ethnic admixture of Jews or non-Samaritans with Greeks would not affect them genetically that much.



Also, there is no reason to assume that the Philistines were genetically Minoan, if anything they'll probably be somewhere between Mycenaeans and Anatolia_EBA.
We won't know until they finish analyze the samples from that cemetery. If they originate in Crete, and arrived to the Levant at around 11th-12th centuries BC, then surely, they would be genetically Minoan. The Minoans completely disappeared as a separate culture around the 15th century, their latest cultural presence would be Crete.
Mycenaeans didn't totally replace them, they just culturally assimilated them.

Agamemnon
06-06-2018, 02:22 PM
I strongly disagree regarding the model for Druze. First, Bedouins are a modern population. Second - why would you add Mycenaean ? on what grounds? Also, Early Bronze Age Armenians and Bronze Age Anatolians highly resemble Aegean and Minoans. Also 25% Bronze Age Caucasians and Anatolians is too high, IMO, and Mycenaean and Minoan admixture gives you much better results.

I was talking about the fits you quoted, not the one I just posted. You're asking the wrong question: Why should we not add Mycenaeans?

EBA Armenians most definitely do not resemble Aegean populations, this is basically Kura-Araxes we're dealing with here.


I wonder, how do people explain admixture of Levantines among modern Greek Islanders, while tend to reject the notion of Greek admixture with Levantines, when in fact Greek colonization in the Eastern Mediterranean was very prominent from at least the 10th century BC, onward.


There's no compelling evidence in favour of Levantine admixture among Aegean Greeks, if that's what you're talking about.


As for the Avars - do you know they are mainly Caucasian and haven't mixed with Turks or Iranians or Eurasian Steppes people in the last +2500 years? Why not use Dagestani? Or Circassians? Because it just gives you a better %% ? Also, this model might be more seasonable than the ones I've quoted which you provided more than a year ago - Samaritans & Bergamo Italians? and 60 / 40 ? Bergamo are highly Celtic. And the result you got 0.31% - is, according to the standards you provide, a very good example of over-fitting. To me, just throwing a bunch of different ethnicities without any historical background, or very far-fetched historical background, is just inaccurate.


I've used Kabardians, Lezgins and Chechens as well, I don't really understand why you're so fixated on the Avars quite frankly. Same thing for Bergamo, at best they derive 25% of their ancestry from a (non-Italian and non-Iberian) Bell Beaker-like source, the rest is Tuscan-like.

Just throwing a bunch of different ethnicities without any historical background, eh? Well, truth be told, that's more or less what your fits look like for the time being.


The only modern ones I care to use, is the Northern European admixture, for a very good reason - according to the study I've quoted in the original post, the admixture with Northern Europeans happened just after the population bottleneck Ashkenazi Jews suffered from - in the last 20 generations. After the 10th century, there were very little mass movements of whole ethnic groups into Western Europe. So basically, German from 750 years ago would resemble genetically a German today. Modern Avars would very likely not resemble BA Armenians and Anatolians.


You don't have much of an excuse here either as we have a lot of ancient samples from Europe, including medieval samples from Germany.


Nope, they're also pretty consistent. Aegean and Levantine admixture is consistent. If I would see in Druze 60% Aegean but Lebanese Christians would have 20%, I would say there is some inconsistency. This is not the case. And Samaritans, according to my model, are indeed pretty much identical to Bronze Age Canaanites (and seem to lack any Aegean admixture).


Not quite, one merely needs to look at the distance of the fits you posted to notice that something just doesn't add up here. It looks like you're trying very hard to produce fits that fit a certain narrative, in this case one of widespread Greek settlement in the Levant.



Samaritans are not the half Jews. I'm talking about Greek "God fearers" and Greek converts which joined Hellenistic Jewish communities all around the East Mediterranean. Or do you deny their existence?

I am merely stating that we should not make much out of such terms, as they were used very liberally by ancient writers.


Actually, I don't know how they couldn't have a demographic impact on the Levant. They have been colonizing it since at least the 10th century BC, and there was even a Minoan settlement found in Northern Israel:

Source: wikipedia's article on Minoan civilization.

And also second wave of Greek colonization:



23709

While indeed, before the 4th century BC, most Greek colonization in East Mediterranean was not in specifically in the Levant (but also existed there), after the 4th century BC, and all the way to the 8th century AD, for more than 1200 years, there have been Greek people coming to the Levant.

After 300 BC, the Hellenistic colonization began. Dark Blue - Greek-majority areas, Pale Blue - highly Hellenized local population, around 300 BC:

23708

And this is what wikipedia has to say:



And:




Come to think about it, it actually makes a lot of sense why Samaritans lack any significant Greek admixture. By the 4th century BC, they were already a closed group of people, separated from Jews. Any ethnic admixture of Jews or non-Samaritans with Greeks would not affect them genetically that much.



We won't know until they finish analyze the samples from that cemetery. If they originate in Crete, and arrived to the Levant at around 11th-12th centuries BC, then surely, they would be genetically Minoan. The Minoans completely disappeared as a separate culture around the 15th century, their latest cultural presence would be Crete.
Mycenaeans didn't totally replace them, they just culturally assimilated them.

We've also found evidence of Minoan presence in Egypt, for some odd reason I don't see you claiming that the Copts derive a substantial portion of their ancestry from a Minoan source... Why is that? As a matter of fact, the Greek presence was arguably far more important in Egypt than it ever was in the Levant.

The Philistines spoke an IE language, arguably an early form of Greek, which they wrote down using the Cypro-Minoan script. This alone makes any theory according to which they were genetically identical to the Minoans very doubtful. Furthermore, you're ignoring the outlier from Armenoi, as well as the fact that the Mycenaeans form a tight cluster despite coming from different parts of Greece.

Erikl86
06-06-2018, 02:40 PM
To everyone asking why re-modelling Bronze Age Canaanite according to Hebar et al. model (Neolithic Levantines + CHL Iranians), instead of using sampled from Bronze Age Levant.

There were series of mass immigrations of Bronze Age Anatolian people, forming empires that controlled the Levant, and mixed with the Bronze Age Levantines, from the 21st century BC, all the way to the 12th century BC.

Hurrians (circa 2300 BC):

23718

Hittites (established their empire circa 1600 BC):

23719

And Mitanni (circa 1400 BC):

23720

And, the Phoenicians themselves started establishing colonies all throughout the Mediterranean around 1200 BC, perhaps mixing with other Mediterranean people.

So, just to be on the safe side, that those samples which are dated from 2000 BC - 1500 BC, are not "contaminated" (bad phrase, I know) with non-Levantine, Indo-European people, and separating the Levantine component from the Mycenaean and Minoan admixture which would also be related to Neolithic Anatolians (so we can see exactly how much is Levantine, and how much is Aegean), which themselves would be close to BA Anatolians to some degree, and might confuse the calculations. I've used the model of Neolithic Levant + CHL Iranian, an mixture which happened around 6000 years ago, long before these empires existed.

Also, as I've said before:

"The Minoans could be modelled as a mixture of the Anatolia
Neolithic-related substratum with additional ‘eastern’ ancestry, but
the other two groups had additional ancestry: the Mycenaeans had
approximately 4–16% ancestry from a ‘northern’ ultimate source related
to the hunter–gatherers of eastern Europe and Siberia (Table 1), while
the Bronze Age southwestern Anatolians may have had ~ 6% ancestry
related to Neolithic Levantine populations. "

Come to think about it, perhaps, the fact that many get high percentage of Minoans, and less so of Mycenaean, might also show that the admixture comes from Aegean people, because Mycenaeans were mixed with Bronze Age Anatolians to a small degree, and Minoans were not, and this might actually be from Bronze Age Hittite, Mitanni or Hurrian admixture with Levantines (so maybe 20% Mycenaean is actually 7% Bronze Age Anatolian, 13% Aegean, for example). But this is just an assumption.

Erikl86
06-06-2018, 03:05 PM
I was talking about the fits you quoted, not the one I just posted. You're asking the wrong question: Why should we not add Mycenaeans?

EBA Armenians most definitely do not resemble Aegean populations, this is basically Kura-Araxes we're dealing with here.

Why would you add Mycenaean? If there was no significant Greek admixture among Levantines?


There's no compelling evidence in favour of Levantine admixture among Aegean Greeks, if that's what you're talking about.
Not ancient, but modern Island Greeks. Also, Modern Cypriot Greeks are heavily admixed with Levantines.


I've used Kabardians, Lezgins and Chechens as well, I don't really understand why you're so fixated on the Avars quite frankly. Same thing for Bergamo, at best they derive 25% of their ancestry from a (non-Italian and non-Iberian) Bell Beaker-like source, the rest is Tuscan-like.

And? What did you get for those? These are all very different people. Modern Caucasus is extremely different from Bronze Age Caucasus. Bell-Beaker like source is too old for any talk about Jews in Italy. Currently, Northern Italians so up north are mixed with Celts (and I would assume Germanic people as well). Also, all of modern Italians are mixed with assimilated Greeks from south Italy after it was completely Latinized in early Medieval times.


Well, truth be told, that's more or less what your fits look like for the time being.

You don't have much of an excuse here either as we have a lot of ancient samples from Europe, including medieval samples from Germany.
Nope, I do not - I use the same admixture of Bronze Age Aegeans and a model of Bronze Age Levantines. Also, for non-Mediterranean admixture, I've used in the Basal-rich K7 French, because I had no stats for Basal-rich K7 results for Medieval Germans, and East French made the most sense, other than Lithuanians alone (which you used as a model. Btw, you got a very low result for Northern European admixture).


Not quite, one merely needs to look at the distance of the fits you posted to notice that something just doesn't add up here. It looks like you're trying very hard to produce fits that fit a certain narrative, in this case one of widespread Greek settlement in the Levant.

I don't really. I just examined a lot of different combinations, including in K36 calculations, and found out that ancient Aegean Greeks give the best fit. Lacking ancient Roman samples, they would also provide a pretty "clean" sample of ancient Mediterranean people that existed during the formation of the Jews as a nation, or close enough to those (I would assume Greeks from the 4th century BC would be pretty similar genetically speaking to Myceneans).

Then, I've tried to find historical justifications for this. I've found that Greek colonization was extremely substantial in the East Mediterranean and also in South Italy, and that happened to be where Western Jews cluster today as well, genetically speaking.

It seems there is enough historical background to support this idea.

And other then insisting that my results do not reflect the reality, without any explanation, you have yet to bust this claim. I suspect your "attack" on this theory is subjective, not objective.


We've also found evidence of Minoan presence in Egypt, for some odd reason I don't see you claiming that the Copts derive a substantial portion of their ancestry from a Minoan source... Why is that? As a matter of fact, the Greek presence was arguably far more important in Egypt than it ever was in the Levant.

Oh but I do ! I do believe Copts should appear to be admixed with Greeks. Actually, I would believe many Copts would probably have higher admixture with ancient Greeks than Christian Lebanese.
I have yet to examine it - because sadly, I don't have their Basal-rich K7 stats.


The Philistines spoke an IE language, arguably an early form of Greek, which they wrote down using the Cypro-Minoan script. This alone makes any theory according to which they were genetically identical to the Minoans very doubtful.
True, we will need to wait for their result. This is just one theory I got. Also, there is also some historical evidence connecting them to Crete - which would also make the case for them being genetically Minoan.

Erikl86
06-06-2018, 03:29 PM
@Agamemnon,

My main claim is very simple, and I'll repeat it.

As a result of ancient (BC) Greek admixture with Judeans, in Judea itself and elsewhere, in the Hellenistic diaspora, and perhaps even before that - as a result of Philistines being absorbed into Judean people (the ones which remained after the Babylonian diaspora), Western Jews were formed. This means, that European admixture among Western Jews, which is mostly Mediterranean, shouldn't be looked at as the result of Jews being expelled from their lands - but something which started in Judea, starting mainly from Hellenistic times.

The reason why modern Western Jews cluster so close with South Italians, Sicilians and Islander Greeks, is because all of these locations have substantial ancient Greek ancestry. Furthermore, as a result of Sicilians being a mixture of ancient Greeks and Levantines (from Phoenician, Punic and then Arabic rule), and a small amount of recent Northern European admixture, they score closest to Western Jews.

I do not reject the notion that some Roman women also joined the admixture - I just say that if they did, it was mainly Greco-Romans who did.

After Muslim occupation of Sicily for 250 years in Early Medieval Ages, many Greco-Sicilians fled to mainland Italy, where they mixed with the rest of the Latin-speaking population. Also, Southern Mainland Italy, which was also Greek-speaking and had Greek majority, was fully Latinized by the 10th century. Muslim raids on mainland southern Italy would drive some of them north as well. This means, that while the most Greek population in Italy is still Sicily and the South, Greek and Levantine admixture exist in one way or another throughout all of the Italian peninsula, less in the North, more in the South. This would further make it look like as if Italians resemble Western Jews.

Also, the last, Northern European admixture, at least for Ashkenazi Jews, was probably the Rhine valley, which has both Celtic and Germanic populations. And north Italy (northern than Tuscany) is also Celtic to some degree, and has higher German influence than the rest of Italy. Which would again, make Italians seem close to Western Jews.

Modern Greek Islanders are closer to Sicilians than to modern Mainland Greeks. Modern mainland Greeks are admixed with Early Medieval Slavs, which entered the Balkans. Modern Greek Islanders have kept isolated from that to a degree, and also got more recent Levantine admixture (which make them appear close to Sicilians and Western Jews, again).

This is my theory, that's it. Nothing conspirative or "Jews do not belong to the Middle East" kind of stuff.

kingjohn
06-06-2018, 06:28 PM
@Agamemnon,

My main claim is very simple, and I'll repeat it.

As a result of ancient (BC) Greek admixture with Judeans, in Judea itself and elsewhere, in the Hellenistic diaspora, and perhaps even before that - as a result of Philistines being absorbed into Judean people (the ones which remained after the Babylonian diaspora), Western Jews were formed. This means, that European admixture among Western Jews, which is mostly Mediterranean, shouldn't be looked at as the result of Jews being expelled from their lands - but something which started in Judea, starting mainly from Hellenistic times.

The reason why modern Western Jews cluster so close with South Italians, Sicilians and Islander Greeks, is because all of these locations have substantial ancient Greek ancestry. Furthermore, as a result of Sicilians being a mixture of ancient Greeks and Levantines (from Phoenician, Punic and then Arabic rule), and a small amount of recent Northern European admixture, they score closest to Western Jews.

I do not reject the notion that some Roman women also joined the admixture - I just say that if they did, it was mainly Greco-Romans who did.

After Muslim occupation of Sicily for 250 years in Early Medieval Ages, many Greco-Sicilians fled to mainland Italy, where they mixed with the rest of the Latin-speaking population. Also, Southern Mainland Italy, which was also Greek-speaking and had Greek majority, was fully Latinized by the 10th century. Muslim raids on mainland southern Italy would drive some of them north as well. This means, that while the most Greek population in Italy is still Sicily and the South, Greek and Levantine admixture exist in one way or another throughout all of the Italian peninsula, less in the North, more in the South. This would further make it look like as if Italians resemble Western Jews.

Also, the last, Northern European admixture, at least for Ashkenazi Jews, was probably the Rhine valley, which has both Celtic and Germanic populations. And north Italy (northern than Tuscany) is also Celtic to some degree, and has higher German influence than the rest of Italy. Which would again, make Italians seem close to Western Jews.

Modern Greek Islanders are closer to Sicilians than to modern Mainland Greeks. Modern mainland Greeks are admixed with Early Medieval Slavs, which entered the Balkans. Modern Greek Islanders have kept isolated from that to a degree, and also got more recent Levantine admixture (which make them appear close to Sicilians and Western Jews, again).

This is my theory, that's it. Nothing conspirative or "Jews do not belong to the Middle East" kind of stuff.

you forget about eastern europe admixture {poland my dear}
achenazi jews score 5% baltic related clusters in living dna
i know it is low but still there and shouldn't be ignore .....

Eihwaz
06-06-2018, 06:55 PM
you forget about eastern europe admixture {poland my dear}
achenazi jews score 5% baltic related clusters in living dna
i know it is low but still there and shouldn't be ignore .....

Let's not forget about tract-length evidence and IBD sharing with Western Slavs. :P

Truth be told, though, I think that someone versed in f3 stats, z-scores, qpAdmix, etc. should tackle the question. I'm definitely doubting that models with ancient aegeans would pass muster that well, especially since they'd have to be rather convoluted. I don't exactly know for sure, of course, since I lack experience with running that stuff (having read some stuff about it on Davidski's blog and some papers).

kingjohn
06-06-2018, 07:09 PM
Let's not forget about tract-length evidence and IBD sharing with Western Slavs. :P

Truth be told, though, I think that someone versed in f3 stats, z-scores, qpAdmix, etc. should tackle the question.

yes but probably you more :D
because you have non jewish west slavic heritage
so i am sure you have more ibd sharing than pure aschenazi ......
but yes the shared segment are usually with ukranians and eastern poles......

Erikl86
06-06-2018, 07:22 PM
you forget about eastern europe admixture {poland my dear}
achenazi jews score 5% baltic related clusters in living dna
i know it is low but still there and shouldn't be ignore .....

The thing is, most of the Ashkenazi Jews sampled were Western Ashkenazi, which although still have some Polish admixture (because Jews moved back from Eastern Europe to Germany, Hungary and Austria in the last 300 years or so). Out of the 14 Ashkenazi Jews used for the average, 11 were Western Ashkenazi, and only 3 were Eastern Ashkenazi.

But I would agree, that Eastern Ashkenazi Jews should have Slavic admixture, and perhaps even Baltic one (through Western Slavs and Lithuanians).

Agamemnon
06-06-2018, 08:32 PM
^^A clear-cut East-West divide between Ashkenazim fails in many aspects to be honest, some "Western Ashkenazim" have some Eastern European admixture while some "Eastern Ashkenazim" have next to none, it isn't as clear cut as some might think it is.

Either way, I've been a vocal proponent of an admixture event between Levantines and Aegean populations (as opposed to a Levantine + Italian model) to explain the ethnogenesis of Western Jewry for several years, until the Mycenaean results made it clear that this was very unlikely. They're just too close to present-day Eastern Mediterraneans for that to be a reasonable explanation. It's a bit as if someone claimed that the Chuvash are unadulterated Yamnayans or that North Africans are entirely derived from a population like IAM, such a model fails for the exact same reasons Western Jews owing most of their ancestry to a Minoan or Mycenaean source fails, namely because it directly contradicts most of the evidence we have (in terms of IBD, uniparental lineages, even medical genetics for that matter).

Speaking of all the Jews aren't Jews propaganda, I think claiming that Jews are just a bunch of Greek converts will become the new hype when the whole Khazar thing dies out.

Erikl86
06-06-2018, 08:57 PM
^^A clear-cut East-West divide between Ashkenazim fails in many aspects to be honest, some "Western Ashkenazim" have some Eastern European admixture while some "Eastern Ashkenazim" have next to none, it isn't as clear cut as some might think it is.


I totally agree. Especially since most "Western" Ashkenazi Jews are descendants from "Ostjuden" from the Pale of Settlement in Eastern Europe which were allowed to immigrate into the Holy Roman Empire and Hapsburg Empires (and France) in the late 18th century onwards.


Either way, I've been a vocal proponent of an admixture event between Levantines and Aegean populations (as opposed to a Levantine + Italian model) to explain the ethnogenesis of Western Jewry for several years, until the Mycenaean results made it clear that this was very unlikely. They're just too close to present-day Eastern Mediterraneans for that to be a reasonable explanation. It's a bit as if someone claimed that the Chuvash are unadulterated Yamnayans or that North Africans are entirely derived from a population like IAM, such a model fails for the exact same reasons Western Jews owing most of their ancestry to a Minoan or Mycenaean source fails, namely because it directly contradicts most of the evidence we have (in terms of IBD, uniparental lineages, even medical genetics for that matter).

In what way are Mycenaean too close to present-day East Meds? modern Greeks are certainly not 100% Mycenaeans. Even the Islanders, while a bit closer to ancient Greeks thanks to isolation, have had some admixture, including late (post-Neolithic) Levantine one.

And also, if we go by that logic, being that Mycenaeans have had little Levantine admixture in them originally, and they are "just too close to present-day Eastern Mediterraneans" as you put it, how else would you explain Western Jews cluster so closely to Greek Islanders and to Sicilians? (And Sicilians and Greek Islanders cluster much closer to each other than mainland Greeks and Greek Islanders)

I have difficulties when some one tries to use modern Mediterranean admixture to try to predict which one was the main contributor to the formation of Western Jews. Europe - East, West and South - has changed considerably in the last 2000 years, and even if we do assume Western Jews left Judea as almost pure Levantines, and met Mediterraneans Europeans in Italy or somewhere else in Southern Europe - that would happen some 2000 years ago. Europe's mass movements of populations ended around 1000-750 years ago - about 1000-1300 years after those Southern Europeans mixed with those Judeans. Assuming modern populations would model those Europeans is unrealistic and could lead to some unexplained situations.



Speaking of all the Jews aren't Jews propaganda, I think claiming that Jews are just a bunch of Greek converts will become the new hype when the whole Khazar thing dies out.
Oh but I didn't claim this at all - I've simply claimed ancient Greek admixture is natural, historical part of the ancient Levant, and as shown from the other non-Jewish Levantine populations I've examined - it exist among them as well. If Jews would have never left the Levant - they would still be admixture of Levantines and ancient Aegean people. Greek colonization of the Levant and the rest of the Mediterranean is as old as Israelite nationhood itself - spanning from the 10th century BC all the way to the 7th century AD - some 1800 years.
And also, Western Jews are still, according to this model, 35-30% Bronze-Age Levantines. That's not nothing.

Erikl86
06-06-2018, 09:11 PM
As for the closeness of modern Mediterraneans to Mycenaean people, check out Eurogenes blog:


Mycenaean
Minoan_Lasithi 0.780±0.044
Srubnaya 0.220±0.044

Distance: 0.909333794


vs

Greek
Iran_ChL 0.090±0.071
Mycenaean 0.478±0.103
Slav_Bohemia 0.432±0.077

Distance: 0.461783732


Sicilian_East
Bell_Beaker_Germany 0.222±0.077
England_Roman_outlier 0.210±0.134
Mycenaean 0.567±0.163

Distance: 0.504442682


Sicilian_West
England_Roman_outlier 0.216±0.121
Mycenaean 0.503±0.135
Unetice 0.281±0.056

Distance: 0.808464904

*England Roman outlier was a Levantine Gladiator found in Britain, dated to Roman times.

So, it seems even modern Greeks (albeit these are from mainland Greece), and less Mycenaean than Sicilians.

Erikl86
06-06-2018, 09:51 PM
Tried to run nMonte with K36oracle. Shortest distance I've ever seen is ~3%.

Check out when I try to model Ashekanzi Jews with modern populations:

[1] "distance%=3.7351 / distance=0.037351"

Ashkenazi_Eastern_Euro

Gr_Andros 35.65
GR_Cyclades 27.95
Samaritan 23.25
German_SouthWest 6.95
PL_SE_Carpathia 6.20

Not surprisingly, Samaritan, which is a good model for BA Levantine, gets "only" 23%, while Aegean Greek Islanders get much more dominance. This is to be expected, as Levantine, post-Bronze Age admixture (as I said, probably starting from Hellenistic times when Greeks heavily colonized the Levant) in Greek Islanders overlaps with Samaritans.

Also, North/Eastern European get's too low presence - this is probably due to small amount of Slavic admixture which penetrated from mainland Greece to the islands, so there's some Slavic overlap there as well.

But overall, the model fits my proposed theory, even with modern populations, in another calculator (although the downsides of modelling ancient admixture events using modern populations is problematic, and it's a good example why).

As for exactly when Greek Islanders got their Levantine admixture, here are some graphs from the 2017 study about Mycenaean, Minoans and modern Greeks and Sicilians:

23728
This is the admixture chart. As you can see, Sicilians and Calabrese are clearly closest to the Cretans and even Dodecanese, and have not only much more of the red "Near East" component than do mainland Greeks.

From the same article, I've cropped the relevant dating of admixture for Southern Balkan area (Greece), which, as the table above show, most of the Levantine admixture was concentrated in Greek Islanders (or, perhaps because there was less Slavic influence there, it is more substantial there).
23729
Circled in red - all Middle Eastern admixture events.

As can be seen from these tables and from the genetic admixture of Mycenaean and Minoan people, it is quite obvious Levantine admixture entered Greek people much later - but is there today. Again, I believe some of it, or at least most of it, started after Hellenistic period, but also perhaps during Byzantine period when Levantines and Greeks shared similar religion and perhaps many Levantines fled the area to relatively calmed Greece during the centuries long Byzantine-Persian war which depopulated and destroyed a lot of the Middle East, setting the grounds for Arabs to take over by the 7th century (which might explain the existence of Maronite Lebanese Cypriots, which date to Medieval ages).

So this is one of the reasons why modern Greeks, and especially Greek Islanders, cluster so closely with Sicilians, which have substantial Greek and Levantine ancestry as well. This is also why they cluster closely to Western Jews. Anyhow, it is quite obvious Mycenaeans aren't the exact copy of modern Greeks.

kingjohn
06-06-2018, 10:00 PM
Tried to run nMonte with K36oracle. Shortest distance I've ever seen is ~3%.

Check out when I try to model Ashekanzi Jews with modern populations:

[1] "distance%=3.7351 / distance=0.037351"

Ashkenazi_Eastern_Euro

Gr_Andros 35.65
GR_Cyclades 27.95
Samaritan 23.25
German_SouthWest 6.95
PL_SE_Carpathia 6.20

Not surprisingly, Samaritan, which is a good model for BA Levantine, gets "only" 23%, while Aegean Greek Islanders get much more dominance. This is to be expected, as Levantine, post-Bronze Age admixture (as I said, probably starting from Hellenistic times when Greeks heavily colonized the Levant) in Greek Islanders overlaps with Samaritans.

Also, North/Eastern European get's too low presence - this is probably due to small amount of Slavic admixture which penetrated from mainland Greece to the islands, so there's some Slavic overlap there as well.

But overall, the model fits my proposed theory, even with modern populations, in another calculator (although the downsides of modelling ancient admixture events using modern populations is problematic, and it's a good example why).

As for exactly when Greek Islanders got their Levantine admixture, here are some graphs from the 2017 study about Mycenaean, Minoans and modern Greeks and Sicilians:

23728
This is the admixture chart. As you can see, Sicilians and Calabrese are clearly closest to the Cretans and even Dodecanese, and have not only much more of the red "Near East" component than do mainland Greeks.

From the same article, I've cropped the relevant dating of admixture for Southern Balkan area (Greece), which, as the table above show, most of the Levantine admixture was concentrated in Greek Islanders (or, perhaps because there was less Slavic influence there, it is more substantial there).
23729
Circled in red - all Middle Eastern admixture events.

As can be seen from these tables and from the genetic admixture of Mycenaean and Minoan people, it is quite obvious Levantine admixture entered Greek people much later - but is there today. Again, I believe some of it, or at least most of it, started after Hellenistic period, but also perhaps during Byzantine period when Levantines and Greeks shared similar religion and perhaps many Levantines fled the area to relatively calmed Greece during the centuries long Byzantine-Persian war which depopulated and destroyed a lot of the Middle East, setting the grounds for Arabs to take over by the 7th century (which might explain the existence of Maronite Lebanese Cypriots, which date to Medieval ages).

So this is one of the reasons why modern Greeks, and especially Greek Islanders, cluster so closely with Sicilians, which have substantial Greek and Levantine ancestry as well. This is also why they cluster closely to Western Jews. Anyhow, it is quite obvious Mycenaeans aren't the exact copy of modern Greeks.

totaly agree modern mainland greeks score 18-25% baltic in eurogenes k13
and many can be modeled in 4 population oracle with ukranian as one of the 4
the oracle need the slavs time and time again ..... :D
the slavs didn't changed the language of the greeks like what happen to the thracians in bulgaria
but they defiantly had genetic impact ......

Bas
06-06-2018, 10:05 PM
I think historically it's definitely plausible but the PCA position of Ashkenazi Jews cannot be explained by the European admixture being Greek. I think even Greek would be a shade too genetically similar to the Levant to have Ashkenazi appearing where they do.

Starting with the obvious Levantine base, they would need something quite a bit more Northern to compensate to be clustering in the same place as Greek and Italian_South. qpAdm runs I did a while back showed the best fit being Italian_North.

John Doe
06-06-2018, 10:21 PM
^^A clear-cut East-West divide between Ashkenazim fails in many aspects to be honest, some "Western Ashkenazim" have some Eastern European admixture while some "Eastern Ashkenazim" have next to none, it isn't as clear cut as some might think it is.

Either way, I've been a vocal proponent of an admixture event between Levantines and Aegean populations (as opposed to a Levantine + Italian model) to explain the ethnogenesis of Western Jewry for several years, until the Mycenaean results made it clear that this was very unlikely. They're just too close to present-day Eastern Mediterraneans for that to be a reasonable explanation. It's a bit as if someone claimed that the Chuvash are unadulterated Yamnayans or that North Africans are entirely derived from a population like IAM, such a model fails for the exact same reasons Western Jews owing most of their ancestry to a Minoan or Mycenaean source fails, namely because it directly contradicts most of the evidence we have (in terms of IBD, uniparental lineages, even medical genetics for that matter).

Speaking of all the Jews aren't Jews propaganda, I think claiming that Jews are just a bunch of Greek converts will become the new hype when the whole Khazar thing dies out.

Ha! If it ever dies, I still see it floating around the internet by people who I don't think if asked can say where Khazaria even was or who the Khazarians actually were, it's funny and sad at the same time

Erikl86
06-06-2018, 10:29 PM
I think historically it's definitely plausible but the PCA position of Ashkenazi Jews cannot be explained by the European admixture being Greek. I think even Greek would be a shade too genetically similar to the Levant to have Ashkenazi appearing where they do.

Starting with the obvious Levantine base, they would need something quite a bit more Northern to compensate to be clustering in the same place as Greek and Italian_South. qpAdm runs I did a while back showed the best fit being Italian_North.

Ashkenazi and Sephardic Jews should originate from the same Mediterranean admixture. It would be difficult to explain North Italian to Sephardic Jews, being that Jewish communities existed all over Italy and Eastern Mediterranean.

I think the problem with PCAs for modern Mediterranean people is that some populations might resemble others because of coincidence. Sicilians cluster with Greeks because substantial amount of Sicilians have Greek ancestry. They also got Levantine and Near Eastern admixture, but different than Greeks (and much older than Greeks). They didn't recieve any Slavic admixture, but got Norman (French/German admixture) admixture.

So you have Mediterranean (Greek + Sardinian-like original Sicilian population) + Levantine/Near Eastern (Phoenician, Punic, Arabic) + North European (Norman). That's pretty much the same admixture for Ashkenazi Jews.

Modern Greeks are Mediterranean (ancient Greek) + Levantine/Middle Eastern (from Hellenistic and Byzantine times) + East European (Slavic, much more on mainland Greece but some of it got to the Islands as well).

And Western Jews, which are Levantines (Bronze Age Canaanites) + Mediterranean (Greek/Roman from Roman and Hellenistic times) + North/East Europeans (southern French for Sephardic Jews, Rhineland Germans and East Europeans for Ashkenazi).

They would cluster similar, but because they have similar genetic admixture, out of pure chance. Should Slavic people not penetrate into the Balkans, then no Slavic admixture for modern Greeks, hence putting them further away from Ashkenazi Jews. No Norman occupation of Sicily and Southern Italy? same deal.

Yet neither Greeks or Sicilians have North Italian admixture as their Mediterranean component - there is no need for it. And yet they all - Sicilians, Greek Islanders and Western Jews cluster very close to each other on PCAs.

Awale
06-06-2018, 10:33 PM
Either way, I've been a vocal proponent of an admixture event between Levantines and Aegean populations (as opposed to a Levantine + Italian model) to explain the ethnogenesis of Western Jewry for several years, until the Mycenaean results made it clear that this was very unlikely. They're just too close to present-day Eastern Mediterraneans for that to be a reasonable explanation. It's a bit as if someone claimed that the Chuvash are unadulterated Yamnayans or that North Africans are entirely derived from a population like IAM, such a model fails for the exact same reasons Western Jews owing most of their ancestry to a Minoan or Mycenaean source fails, namely because it directly contradicts most of the evidence we have (in terms of IBD, uniparental lineages, even medical genetics for that matter).

Indeed, I got to witness your change in perspective in real-time. I still recall how often you'd mention that Greeks were more likely than Italics to be the source of the non-Levantine ancestry in Western-Jews (well, most of the source) yet you completely abandoned that after the Mycenaean samples came out because they appeared too similar to modern East-Meds already. Plus, even with current models using G25 and nMonte3 (pen=0 using scaled data) they don't seem to produce very convincing results to me:

distance%=2.0665

Ashkenazi-Jew

Samaritan 48.2
Mycenaean 26.6
Polish 25.2


distance%=2.3582

Sephardic-Jew

Samaritan 52
Mycenaean 33.2
Polish 14.8

That's waaay too much Polish right there in the Ashkenazim and the Sephardim shouldn't be showing Polish at all. No such issues when I plug in Tuscans, though:

distance%=1.6083

Ashkenazi-Jew


Samaritan 45.2
Tuscan 45
Polish 9.8

distance%=2.0149

Sephardic-Jew

Samaritan 51.4
Tuscan 48.6

I assume the problem is that they have too little Steppe ancestry to make sense. But, I don't think one can totally rule out later Greeks who'd possibly have more steppe ancestry. We'd have to see Classical samples but I have a feeling even they won't fit as well as Tuscans do because modern Greeks from areas like Macedonia or Thessaly, despite their Slavic admixture, don't cluster more north than Tuscans from what I remember.

coffeeprince
06-06-2018, 10:38 PM
That was an interesting read, Erik. Thank you for taking the time to write it.
My Y haplogroup may have originated in Greece (or Crete, possibly Minoans or Mycenaean), but I'll need to further test to find out.

Erikl86
06-06-2018, 10:54 PM
Indeed, I got to witness your change in perspective in real-time. I still recall how often you'd mention that Greeks were more likely than Italics to be the source of the non-Levantine ancestry in Western-Jews (well, most of the source) yet you completely abandoned that after the Mycenaean samples came out because they appeared too similar to modern East-Meds already. Plus, even with current models using G25 and nMonte3 (pen=0 using scaled data) they don't seem to produce very convincing results to me:

distance%=2.0665

Ashkenazi-Jew

Samaritan 48.2
Mycenaean 26.6
Polish 25.2


distance%=2.3582

Sephardic-Jew

Samaritan 52
Mycenaean 33.2
Polish 14.8

That's waaay too much Polish right there in the Ashkenazim and the Sephardim shouldn't be showing Polish at all. No such issues when I plug in Tuscans, though:

distance%=1.6083

Ashkenazi-Jew


Samaritan 45.2
Tuscan 45
Polish 9.8

distance%=2.0149

Sephardic-Jew

Samaritan 51.4
Tuscan 48.6

I assume the problem is that they have too little Steppe ancestry to make sense. But, I don't think one can totally rule out later Greeks who'd possibly have more steppe ancestry. We'd have to see Classical samples but I have a feeling even they won't fit as well as Tuscans do because modern Greeks from areas like Macedonia or Thessaly, despite their Slavic admixture, don't cluster more north than Tuscans from what I remember.

If you use Mycenaean samples, why use Samaritans and not re-model Bronze Age Levantines by mixing Neolithic Levantines and CHL Iranian samples?

There is no logic in that. You get screwed up results, for a very good reason:
Bronze Age Anatolian admixture in Samaritans is guaranteed. I believe it even exist in BA Sidon samples themselves. As I've explained in this thread, Bronze Age Anatolian Empires have been invading and rulling the Levant from the 21st century BC all the way to the 12th century BC (Hurrians, Hittites, Mitanni).

It is very possible modern Samaritans overlap somewhat with Mycenaean, because of that Anatolian Bronze Age admixture which happened to the Canaanites which Samaritans originate from.

I would recommend re-constructing the Hebar Bronze Age Levantine for "cleaner" analysis.

I also agree with your assumption that perhaps the Steppe ancestry in Mycenaean caused Polish to appear for Sephardic Jews. That was my assumption as well.

Modern Tuscans are problematic because they would be more biased - they would have some, albeit very small, Levantine admixture from Southern Italians and Sicilians immigrating North over the years, and also some Greek ancestry for the exact same reasons.

Plus, Tuscans would already include some (small) amount of northern European (Celtic) admixture that would satisfy Sephardic Jews. Also, 9.8% Polish for Ashkenazi Jews is very small North/Eastern admixture, and obviously there's overlap of some sort with Tuscans.

Agamemnon
06-06-2018, 11:01 PM
Ha! If it ever dies, I still see it floating around the internet by people who I don't think if asked can say where Khazaria even was or who the Khazarians actually were, it's funny and sad at the same time

Oh but it will die, slowly but surely, make no mistake about it. Antisemitism is evolving, the main focus now is about de-Judaising the Jews, something the Nazis themselves didn't really engage in (they were content with labeling Jews "untermenschen"). As the discoveries which are now common knowledge on this forum filter down to the public, and they inevitably will, we will see a shift from "Jews are just Khazars" (which is rampant right now) to "Jews are just Greek converts". Think of how the Rothschild jewspiracies are slowly fading while Soros jewspiracies are gaining ground, you still find a few folks here and there talking about the Rothschilds and the central bank but all the hype is around Soros (and understandbly so, as he's a very opaque fellow). Some of the main hives of antisemitism still aren't receptive to the world of population genetics, which they perceive as a Jewish pseudoscience, but this will also change pretty fast.

It isn't hard to imagine a future where we'll miss the Khazar nonsense, nothing quite says "I'm an idiot" like claiming that an Eastern Mediterranean population is actually descended from Lir-Turkic nomads from the Caspian steppe, the idiocy is transparent here. Claiming that Jews are just a bunch of Greek converts is trickier however since at first glance, this is a plausible theory, at some level this means you are familiar with the genetic data.

Anyway, sorry for the rant, I know this is off-topic but still, the evolution of antisemitism is an interesting topic in itself and as someone who keeps a very close eye on what antisemites say (I've literally written down hundreds of antisemitic comments which caught my eye since last year) I couldn't help but answer.


I assume the problem is that they have too little Steppe ancestry to make sense. But, I don't think one can totally rule out later Greeks who'd possibly have more steppe ancestry. We'd have to see Classical samples but I have a feeling even they won't fit as well as Tuscans do because modern Greeks from areas like Macedonia or Thessaly, despite their Slavic admixture, don't cluster more north than Tuscans from what I remember.

Indeed, but keep in mind that most of the arguments for keeping the Greeks as a potential source of admixture are basically special-pleading. The only way this is going to work is if we assume Armenoi wasn't really an outlier and that this individual is representative of what post-LBA collapse Greece was like.

The main issue with this is that a "Doric invasion" is often invoked, this fails on many accounts because the traditionally Doric-speaking areas of Greece are inhabited by the populations which are the closest to the Mycenaeans in all of Greece. More importantly, the Mycenaean samples, despite a relatively low sample size, come from a wide area, from the Western Peloponnese to Attica, and show minimal variation over this large area, so it's unlikely we'll see much substructure in upcoming results at this stage.

Erikl86
06-06-2018, 11:18 PM
Samaritan Bronze Age Anatolian admixture:

nMonte3:
[1] "distance%=1.4027"

Samaritan:GSM537032

Levant_BA,80.6
Anatolia_BA,19.4


nMonte2:
distance%=1.2936 / distance=0.012936"

Samaritan:GSM537032

Levant_BA:I1706 79.3
Anatolia_BA:I2495 20.6

Had to use Levant_BA, because Anatolian BA include substantial CHL_Iranian, which would cause overlap in re-consturcting the model in case of trying to model such "archaic" population as Samaritans.

Let's try to see the distance for just Levant_BA, without Anatolian BA samples:

nMonte2:
distance%=3.2376 / distance=0.032376"

Samaritan:GSM537032

Levant_BA:I1705 100
Levant_BA:I1706 0
Levant_BA:I1730 0

nMonte3:
[1] "distance%=3.2376"

Samaritan:GSM537032

Levant_BA,100

So as was predictable by history of the region, modern Samaritans do have BA Anatolian admixture, about 20% of it (maybe even CHL_Anatolians would make a better fit, as Hittites and other MBA Anatolians had substantial CHL Anatolian component), and 80% Bronze Age Canaanite component. That would then explain why using Myceneans, which also had substantial BA Anatolian admixture, and modern Samaritans, would give such freaky results.

Again, IMO, only recent (less than 1000 years) Northern/Eastern European populations could be used is these calculations, because that's when the mass migrations of people have stopped in Europe and a German or a Pole from 1250 AD is pretty much, genetically speaking, similar to a 2018 AD German or Pole (of course, some genetic drift and other things still play a role). For admixtures which occurred before that, we should use ancient samples, and be wise about it.

Erikl86
06-06-2018, 11:23 PM
Oh but it will die, slowly but surely, make no mistake about it. Antisemitism is evolving, the main focus now is about de-Judaising the Jews, something the Nazis themselves didn't really engage in (they were content with labeling Jews "untermenschen"). As the discoveries which are now common knowledge on this forum filter down to the public, and they inevitably will, we will see a shift from "Jews are just Khazars" (which is rampant right now) to "Jews are just Greek converts". Think of how the Rothschild jewspiracies are slowly fading while Soros jewspiracies are gaining ground, you still find a few folks here and there talking about the Rothschilds and the central bank but all the hype is around Soros (and understandbly so, as he's a very opaque fellow). Some of the main hives of antisemitism still aren't receptive to the world of population genetics, which they perceive as a Jewish pseudoscience, but this will also change pretty fast.

It isn't hard to imagine a future where we'll miss the Khazar nonsense, nothing quite says "I'm an idiot" like claiming that an Eastern Mediterranean population is actually descended from Lir-Turkic nomads from the Caspian steppe, the idiocy is transparent here. Claiming that Jews are just a bunch of Greek converts is trickier however since at first glance, this is a plausible theory, at some level this means you are familiar with the genetic data.

Anyway, sorry for the rant, I know this is off-topic but still, the evolution of antisemitism is an interesting topic in itself and as someone who keeps a very close eye on what antisemites say (I've literally written down hundreds of antisemitic comments which caught my eye since last year) I couldn't help but answer.
.

Who claims Jews are a bunch of Greek converts?? Obviously, we are not. Obviously, there is substantial Levantine component in Western Jews. If anything, this just shed light on the fact that many modern Levantines are also mixed with ancient Greek ancestry, because my assumption is that this admixture happened here, in the Levant, instead of Western Jews forming in Roman Tuscany.

How come in your mind, Western Jews being 50% European, with all of that European admixture happening outside of the Levant, is OK, but a plausible theory that shows how this could have happened in the Levant itself, and it's intimidate surrounding areas, seem to you like de-legitimizing Jewish connection to the Levant?

And please, I'm Jewish as well - no need for anti-Semitic rant and evoking Nazis for god sake :-/.

Agamemnon
06-06-2018, 11:33 PM
Who claims Jews are a bunch of Greek converts?? Obviously, we are not. Obviously, there is substantial Levantine component in Western Jews. If anything, this just shed light on the fact that many modern Levantines are also mixed with ancient Greek ancestry, because my assumption is that this admixture happened here, in the Levant, instead of Western Jews forming in Roman Tuscany.

How come in your mind, Western Jews being 50% European, with all of that European admixture happening outside of the Levant, is OK, but a plausible theory that shows how this could have happened in the Levant itself, and it's intimidate surrounding areas, seem to you like de-legitimizing Jewish connection to the Levant?

And please, I'm Jewish as well - no need for anti-Semitic rant and evoking Nazis for god sake :-/.

Rest assured, I am not even remotely implying that you are an antisemite (which would be ridiculous), in fact my rant had nothing to do with you or your theory to begin with. I am merely discussing the changes we're probably going to see in the near future in terms of antisemitic discourse.

Regarding your question: I think the most unlikely part of the scenario you're putting forth is the implication that the admixture event occurred in the Levant. If that were the case, you'd at least expect to see the same Aegean component in non-Western Jewish populations, which obviously are of Jewish origin (though in their case, the bulk of their ancestry does indeed seem to be derived from local "converts"). But that's not what we see, instead we clearly see a Samaritan-like component both in Mizrahi and Yemeni Jews. From a historical standpoint, the admixture event is more likely to have taken place either in the Italian peninsula (probably Rome and its immediate surroundings, if not the northern parts of the Italian peninsula) or, if we assume that your scenario is correct, somewhere in the Aegean (where the diaspora was present early on and was numerically substantial), and this admixture event would've primarily been between a pristinely Judean population and a Southern European one. Testing the remains from the Jewish catacombs of Rome would prove very informative.

kingjohn
06-07-2018, 06:56 AM
Rest assured, I am not even remotely implying that you are an antisemite (which would be ridiculous), in fact my rant had nothing to do with you or your theory to begin with. I am merely discussing the changes we're probably going to see in the near future in terms of antisemitic discourse.

Regarding your question: I think the most unlikely part of the scenario you're putting forth is the implication that the admixture event occurred in the Levant. If that were the case, you'd at least expect to see the same Aegean component in non-Western Jewish populations, which obviously are of Jewish origin (though in their case, the bulk of their ancestry does indeed seem to be derived from local "converts"). But that's not what we see, instead we clearly see a Samaritan-like component both in Mizrahi and Yemeni Jews. From a historical standpoint, the admixture event is more likely to have taken place either in the Italian peninsula (probably Rome and its immediate surroundings, if not the northern parts of the Italian peninsula) or, if we assume that your scenario is correct, somewhere in the Aegean (where the diaspora was present early on and was numerically substantial), and this admixture event would've primarily been between a pristinely Judean population and a Southern European one. Testing the remains from the Jewish catacombs of Rome would prove very informative.


probably italy rather than greece don't forget your father 40% bergamo :)

Idwaajeden
06-07-2018, 07:57 AM
Indeed, I got to witness your change in perspective in real-time. I still recall how often you'd mention that Greeks were more likely than Italics to be the source of the non-Levantine ancestry in Western-Jews (well, most of the source) yet you completely abandoned that after the Mycenaean samples came out because they appeared too similar to modern East-Meds already. Plus, even with current models using G25 and nMonte3 (pen=0 using scaled data) they don't seem to produce very convincing results to me:

distance%=2.0665

Ashkenazi-Jew

Samaritan 48.2
Mycenaean 26.6
Polish 25.2


distance%=2.3582

Sephardic-Jew

Samaritan 52
Mycenaean 33.2
Polish 14.8

That's waaay too much Polish right there in the Ashkenazim and the Sephardim shouldn't be showing Polish at all. No such issues when I plug in Tuscans, though:

distance%=1.6083

Ashkenazi-Jew


Samaritan 45.2
Tuscan 45
Polish 9.8

distance%=2.0149

Sephardic-Jew

Samaritan 51.4
Tuscan 48.6

I assume the problem is that they have too little Steppe ancestry to make sense. But, I don't think one can totally rule out later Greeks who'd possibly have more steppe ancestry. We'd have to see Classical samples but I have a feeling even they won't fit as well as Tuscans do because modern Greeks from areas like Macedonia or Thessaly, despite their Slavic admixture, don't cluster more north than Tuscans from what I remember.

Why not Sephardim can show Polish in your "population sharing" models, or as you said none at all?

Ok I see you "assume" some steppe component is causing inflation

Here is my friend data he is actually 100% Polish

Erikl86
06-07-2018, 09:25 AM
Rest assured, I am not even remotely implying that you are an antisemite (which would be ridiculous), in fact my rant had nothing to do with you or your theory to begin with. I am merely discussing the changes we're probably going to see in the near future in terms of antisemitic discourse.

Regarding your question: I think the most unlikely part of the scenario you're putting forth is the implication that the admixture event occurred in the Levant. If that were the case, you'd at least expect to see the same Aegean component in non-Western Jewish populations, which obviously are of Jewish origin (though in their case, the bulk of their ancestry does indeed seem to be derived from local "converts"). But that's not what we see, instead we clearly see a Samaritan-like component both in Mizrahi and Yemeni Jews. From a historical standpoint, the admixture event is more likely to have taken place either in the Italian peninsula (probably Rome and its immediate surroundings, if not the northern parts of the Italian peninsula) or, if we assume that your scenario is correct, somewhere in the Aegean (where the diaspora was present early on and was numerically substantial), and this admixture event would've primarily been between a pristinely Judean population and a Southern European one. Testing the remains from the Jewish catacombs of Rome would prove very informative.

Not necessarily - if most admixture happened after the Babylonian captivity in the 6th century BC, then you won't see much evidence of it among non-Western Jews - which I'd agree, also seem to originate from Judean and Mesopotamian converts.
Until the 6th century BC, there are records of Philistine cities existing as separate entities. After the 6th century, the Philistines disappear as distinctive people in history. So, given that most Mizrahi communities originate from Babylonian diaspora, if the main admixture with Philistines happened with the remaining Judean population - it won't appear in Mizrahi Jews.
But even if we somehow discard Philistine admixture, or if Philistines were already highly mixed with Canaanites so to have almost no distinct Aegean component by the 6th century BC, still, the major arrival of Greek colonialists to the Levant happened after the beginning of the Hellenistic periods, around the 4th century BC, long after the forefathers of Mizrahi Jews have had already established a thriving Jewish community in Babylon.
My claim is that the admixture started out in the Levant, around the 3rd century, where Hellenistic Judaism formed.
These Hellenistic Jews then intermarried with mainly Greek converts which joined the Hellenistic communities throughout all over the Eastern Mediterranean, culminating in the 1st century AD (just before the destruction of the 2nd temple) to this:

23744

The vast majority of Hellenistic Jews then became the first Christians, which became even more popular because early on they cancelled the need for circumcision, which opened the way for many men to convert (before that, most fully converts were women - this is historically documented, as well as genetically).

What would be a good indication for this, is if modern day Christian population in the Levant would also be highly mixed with ancient Aegeans. Of course, they would be more Levantine than Western Jews, because the Western Jews are not just descendants of post-Hellenistic Levantine Jews, but also of diaspora converts.

And after one of the members here asked, I actually did go and check, and found that this is indeed the case:


Let's see about Christian Lebanese:

[1] "distance%=0.0292 / distance=0.000292"

Lebanese_Christian:average

Minoan_Lasithi:I0070 31.55
Iran_ChL:I1661 25.15
Levant_N:I1704 24.50
Iran_ChL:I1665 12.55
Mycenaean:I9041 4.30
Levant_N:I1699 1.65
Iran_ChL:I1662 0.30

Roughly 65% Bronze Age Levantines, 35% ancient Aegean. My bet, these are Greek Orthodox Christian Lebanese which have been sampled

Distance of 0.0292%.

Let's check Samaritans, with the exact same populations as Christian Lebanese:

[1] "distance%=0.5681 / distance=0.005681"

Samaritan:GSM537032

Iran_ChL:I1665 34.5
Levant_N:I1704 33.2
Levant_N:I0867 32.4

No Aegean admixture appeared.

Let's check Christian Israeli Arabs:

[1] "distance%=0.134 / distance=0.00134"

Arab_Israel_1:average

Levant_N:I1704 30.2
Minoan_Lasithi:I0073 26.6
Iran_ChL:I1665 21.8
Iran_ChL:I1662 15.9
Iran_ChL:I1661 5.5

Roughly 73% Bronze Age Levantines, 26% Aegean - mostly Minoan (No Mycenaean).

Next, let's check Druze:

[1] "distance%=0.0289 / distance=0.000289"

Druze:average

Minoan_Lasithi:I0073 21.20
Levant_N:I1704 20.35
Iran_ChL:I1665 19.65
Iran_ChL:I1661 14.40
Mycenaean:I9010 9.15
Iran_ChL:I1662 8.05
Mycenaean:I9006 6.50
Iran_ChL:I1674 0.70

Very similar to Christian Lebanese - Bronze Age Levantine is 63%, Ancient Aegean is 37% (both Minoan and Mycenaean). Distance is 0.0289%.

Now let's check Cypriots:

[1] "distance%=0.0998 / distance=0.000998"

Cypriot:average

Minoan_Odigitria:I9131 24.80
Mycenaean:I9041 21.40
Minoan_Lasithi:I0073 17.10
Iran_ChL:I1661 12.00
Iran_ChL:I1662 8.65
Iran_ChL:I1665 7.25
Levant_N:I1704 6.70
Iran_ChL:I1674 2.10

Mirror image of Druze . 63% Aegean, 37% Levantines. Not surprising at all.


It would actually make sense that Christian Palestinians would be less Greek than Lebanese Christians, because Lebanese Christians' second largest sect is Greek Christians, which probably have more late, direct substantial Greek ancestry (probably during Byzantine period).

Btw, I've read that one of the reasons Ashkenazi Jews and Samaritans are so difficult to accurately model in all of those calculators, is that they are extremely endogenic communities, and tend to have their own genetic drift which make them unique. Meaning if we'll take an Ashkenazi Jew remains from, let's say 1000 AD (the genetic bottleneck seem to have happened after that), and try to see if it clusters with modern Ashkenazi Jews, he might show further away. So, good Western Jews to model and work around would be Sephardic Jews, which anyhow cluster extremely close to Ashkenazim and basically share the same origins (just that Ashkenazi Jews mixed with more North/East Europeans).

Btw, judging how Cypriots actually resemble the model I got for Sephardic Jews w/ ancient Aegeans and BA Levant, that is - ~60% Aegean, ~30% Levantine, I'm going to try to model Sephardic Jews, with only modern populations, using Cypriots. Should be interesting.

Erikl86
06-07-2018, 09:58 AM
And here are the results, using ONLY modern populations. I've used the average of Turkish Sephardic Jews.

Just equating to Cypriots:

distance%=3.3042 / distance=0.033042"

Sephardic_Jew:average

Cypriot:Cyprus22AJ19 100
Cypriot:Cyprus21AJ19 0
Cypriot:Cyprus24AJ19 0

The results are close, but not that much. I've shown here better ones, usually when adding Southern French.

So now, let's add southern French but also North African admixture (which probably exist more among Sephardic Jews than Ashkenazim because of living side-by-side with Moors in Spain for 700 years):

[1] "distance%=0.0486 / distance=0.000486"

Sephardic_Jew:average

Cypriot:Cyprus22AJ19 52.2
Cypriot:Cyprus21AJ19 32.0
French_South:SouthFrench3326 8.5
French_South:SouthFrench3947 4.5
Moroccan:MCA37 1.7
Cypriot:Cyprus24AJ19 1.1

The best results by far ! 85% Cypriot, 13% Southern French, less than 2% West North African. Distance: 0.0468% !

Now, if we are adding Samaritans to the mix, we still get good results, but we get weird dominance of the Southern French component:

[1] "distance%=0.0687 / distance=0.000687"

Sephardic_Jew:average

Cypriot:Cyprus21AJ19 25.75
Cypriot:Cyprus22AJ19 22.05
French_South:SouthFrench3947 20.90
Samaritan:GSM537032 19.15
Cypriot:Cyprus24AJ19 7.00
Turkish:Turkey1 3.40
Moroccan:MCA37 1.75

As can be seen, I've also tried Turkish component, to see if there is any recent Anatolian admixture. It doesn't show up if I don't add Samaritans, and it does if I do, probably again, because adding Samaritans screws up a little bit the results.

Taking out South French completely screws up the results, but look how it also zeros down Samaritan for some reason:

[1] "distance%=3.209 / distance=0.03209"

Sephardic_Jew:average

Cypriot:Cyprus22AJ19 96.35
Turkish:Turkey3 3.65
Cypriot:Cyprus21AJ19 0.00
Cypriot:Cyprus24AJ19 0.00
Moroccan:MCA37 0.00
Tunisian:Tunisian20B4 0.00
Turkish:Turkey1 0.00
Turkish:Turkey2 0.00
Turkish:Turkey4 0.00
Turkish:Turkey5 0.00
Samaritan:GSM537032 0.00

Which kind of proves that Samaritan Levantine admixture is overlapping with Cypriot Levantine admixture, IMO.


My conclusions from all of these results, and if I look at the previous results using ancient Aegean populations, is that the Levantine component in Cypriots is indeed very old - Phoenician perhaps (maybe even Neolithic?), which would make them overlap with Samaritans. Samaritans are genetically isolated enough to perhaps drift away a little bit, so adding them to the mix kind of screws up the results a little. And Southern French seem to be the "Northern" shared component for Sephardic Jews, because without it, the results get completely screwed up.

For me, this is another confirmation for my theory, this time using only modern populations, that Western Jews are indeed ~30% ancient Levantines, ~50-60% Aegean (Basically, make up of modern Cypriots), ~10-20% Northern European. Most likely France is the place of divergence between Sephardic Jews and Ashkenazi Jews - with Ashkenazim getting more Rhineland admixture, then later on Slavic, and Sephardim getting more Southern French, then perhaps a little bit of Moorish admixture seeing how they need Moroccan to get such good results.

The results are worse than what I got when I actually used ancient Aegean samples (~0.01% vs. ~0.04% distance), I would assume for several reasons:
1. Cypriots have genetically drifted from their Aegean origins. Or the amount of Mycenaean vs. Minoan is different from the actual Aegean ancestry of Western Jews.
2. The Levantine admixture in Cypriots is different from the actual Levantine origin of Jews - that would make sense. Either geographic - even Phoenicians vs. Judeans would cause some difference (despite the two being both Canaanite Levantines), or chronological (Jews descend from Bronze Age Levantines, while Cypriots from more Neolithic Levantines with different amount of CHL Iranian admixture because of the isolation on an island).
3. Another small admixture in Cypriots history which might change their stats a little bit (for example - there was some Armenian immigration to Cyprus during Byzantine period).

I will try to model Ashkenazim later on with only modern populations, using this model (perhaps discarding North African component as it shouldn't exist in Ashkenazim, maybe only very small amount because of some Sephardic Jews joining pre-existing Ashkenazi communities after 1492, but my guess is that it would be too little to notice), and trying different kind of North and Eastern European populations.

But, because of their extreme endogenic nature, my guess is that even if I'll hit the spot with the "exact" combination, I would never get such small distance as Sephardic Jews get, because of major genetic drift.

Erikl86
06-07-2018, 11:36 AM
Here are the results for Ashkenazi Jews, using only modern populations.

First of all, because I know both Cypriots and Sicilians and Greek islanders all cluster close with Ashkenazim, I had to test to see if these high results I get with Cypriots is not just because of PCA similarities, so I replaced Cypriots with the closest populations everyone talk about - Sicilians. I've left North and Eastern European admixture.

Surely, if indeed the high admixture of Cypriots is because a PCA of 50% Levantine, 30% North Italian, and 20% North European falls in Greek/Cyprus/Southern Italy, then I would be able to simply replace them with Sicilians in my admixture and expect the same (maybe even better) results.

Let's see:

[1] "distance%=2.094 / distance=0.02094"

Ashkenazi_Jew:average

Italian_South:SouthItalian10H 100
French_East:French24178 0
French_East:French24433 0
French_East:French24434 0
Polish: Poland4 0
Polish: Poland5 0
Polish: Polish10H 0
Polish: Polish7H 0
Polish: Polish8H 0
Italian_Sicilian_East:EastSicilian1H 0
Italian_Sicilian_East:EastSicilian4H 0
Italian_Sicilian_East:EastSicilian9H 0
Italian_Sicilian_West:WestSicilian10H 0
Italian_Sicilian_West:WestSicilian7H 0
Italian_Sicilian_West:WestSicilian8H 0
Italian_South:SouthItalian16H 0
Italian_South:SouthItalian3H 0

Obviously, bogus results. This proves my assumption that the only reason Sicilians and Italians get clustered so close to Jews, is because of coincidence, or they would be replaceable as the south European component for Ashkenazi Jews, like Cypriots are. It's just that substantial Aegean ancestry (as in South Italians and Sicilians) + Levantine ancestry (from Phoenician, Punic and Arab admixture in S. Italians and Sicilians) + North European ancestry (from Normans mostly and to some very small degree - Lombard/Gothic invasions in the case of Italians) gives a very similar model to Jews.

Now, let's try with similar model to what I've used for Sephardic Jews. It should be basically similar, because both originate from the same "proto" Western Jewish population.

So, just adding Eastern European populations give some close results:

[1] "distance%=0.2075 / distance=0.002075"

Ashkenazi_Jew:average

Cypriot:Cyprus21AJ19 47.8
Cypriot:Cyprus22AJ19 30.4
French_South:SouthFrench3947 14.6
Ukrainian_West:UkrLv240 6.5
Polish: Polish8H 0.7

But, northern European admixture is a little bit too high (plus, Ukrainian should be lower than Polish, IMO, but I guess Western Ukrainians are heavily admixed with Eastern Poles and should be looked at as very close population).

So, let's try to take out Southern French, and replace them with Eastern French, and let's also add Germans:

[1] "distance%=0.1816 / distance=0.001816"

Ashkenazi_Jew:average

Cypriot:Cyprus22AJ19 65.10
Cypriot:Cyprus21AJ19 17.85
German:Germany3 11.60
Ukrainian_West:UkrLv240 5.45

Seems better, plus the amount of Northern European admixture makes much more sense. Notice how Germans take precedence over East French. This is ok - it probably means the original Rhineland inhabitants were more Germanic back then.

Next, let's try and add Samaritans to the exact same mix:

[1] "distance%=0.1964 / distance=0.001964"

Ashkenazi_Jew:average

Samaritan:GSM537032 34.4
Cypriot:Cyprus22AJ19 31.1
French_East:French24434 12.2
Ukrainian_West:UkrLv240 10.3
German:Germany3 6.2
Cypriot:Cyprus21AJ19 5.9

Again, just as in the case of Sephardic Jews, adding Samaritans completely screws up the results, while still retaining a pretty good distance. According to this, Ashkenazi Jews are almost 29% Northern Europeans ! Which is of course wrong. By the way, again see how Samaritans, which represent the closest population to Bronze Age Levantines, never get higher than 30% - just like in my model with ancient Aegeans.

I've even tried to take out the Germans, to see if it'll sort out this problem, while still retaining the Samaritans:

[1] "distance%=0.197 / distance=0.00197"

Ashkenazi_Jew:average

Cypriot:Cyprus22AJ19 33.90
Samaritan:GSM537032 32.75
French_East:French24434 16.05
Ukrainian_West:UkrLv240 11.55
Cypriot:Cyprus21AJ19 5.15
Polish: Polish8H 0.60

Still, same bogus results.

Next, I've took out the Samaritan populations, and tried to put Northern Italians. I've also returned East French. Surely, if Italian is the most significant southern European component, as many here claim, they would grab their share of percentage from Cypriots. Let's see:

[1] "distance%=0.2018 / distance=0.002018"

Ashkenazi_Jew:average

Cypriot:Cyprus21AJ19 39.5
Cypriot:Cyprus22AJ19 37.5
Italian_Bergamo:HGDP01177 9.8
Ukrainian_West:UkrLv240 9.4
French_South:SouthFrench3947 3.9


Well well - Cypriots "took a hit", but is still around 77% - haven't changed a bit. But Bergamo got "only" ~10%, and on the expense of.... East French. Which completely disappeared. This would agree with the fact that Northern Italians have substantial Celtic and even some Germanic make up - like East French. Also, southern, perhaps Greek-derived admixture from S. Italy in Bergamo Italians probably what lowered by 2-3% the percentage Cypriots got.

I think it's pretty safe to say, that modern Cypriots, which have a major ancient Aegean ancestry, are much closer genetically to Western Jews than either modern South Italians or modern North Italians.

The only reason, so it seems, that Western Jews cluster close to S. Italians and Sicilians, is because those have substantial Greek ancestry and Levantine admixture. But their Italian component puts them further away from Western Jews, which got most of their Mediterranean admixture from ancient Aegeans.

Btw, not surprising, and probably due to genetic drift caused by the major genetic bottleneck Ashkenazi Jews experienced, it would be very difficult to get the same 0.00X% that I got with Sephardic Jews, especially using modern populations as representative of ancient admixture.

So best fit for Ashkenazi Jews, currently, is:
Cypriots (which are basically 37% Levantines, 63% Aegeans): ~83%
North/Eastern European, with the main component being German, and minor West Ukrainian: 17%.
Distance: 0.18%.

~37% out of 83% Cypriot = 30% Levantine admixture for Ashkenazi Jews, consistence with what I got throughout all of this research.

If I compare that to the best result I got in Sephardic Jews using modern populations, then it would be 37% of 85% Cypriot = ~31.5% Levantine. Same as Ashkenazi Jews, which would be consistence with the same ethnogenesis.

This is also happen to be the same percentage Samaritans, which again are the closest living relatives of Bronze Age Canaanites, genetically speaking, get in both when I add them into the mix (although it screws up the results because of probable overlap with the Levantine component of Cypriots).

Agamemnon
06-07-2018, 12:29 PM
You'd certainly expect to see that same Aegean component among Yemeni Jews at the very least if we were to assume your theory is correct, and that's not what we're seeing.

I can model my mother as 100% Bell Beaker, I can also model her as a mixture of England_Roman and England_IA. Am I to conclude that my mother is some sort of Bronze Age or Iron Age relic? No, because while such a mixture provides a good statistical fit, it doesn't fit with what we know about the British Isles' demographic history. Same thing here with your models including Cypriots (or any other non-Jewish Eastern Mediterranean population for that matter).

kingjohn
06-07-2018, 12:48 PM
use tuscan instead of bergamo aschenazis show only 3-4% north italian in living dna
they show much higher south italy and tuscany
regards
adam

p.s
it is cool to see the southern french in your Sephardi run again it is inline with what i see in full sefhardi results gedmatch , gencove , and dna land

Erikl86
06-07-2018, 01:52 PM
You'd certainly expect to see that same Aegean component among Yemeni Jews at the very least if we were to assume your theory is correct, and that's not what we're seeing.



Aren't Yemenite Jews descendant mainly from Babylonian diaspora? They also show some admixture with local Arab converts, probably from Himyarite kingdom (although paternally, they cluster with the rest of the Jews, showing again how Jewish males from the Levant took local converts).

I haven't tested them, but if their community is mainly of Babylonian of origin, then again - they will not show significant (if any) Aegean ancestry.


I can model my mother as 100% Bell Beaker, I can also model her as a mixture of England_Roman and England_IA. Am I to conclude that my mother is some sort of Bronze Age or Iron Age relic? No, because while such a mixture provides a good statistical fit, it doesn't fit with what we know about the British Isles' demographic history. Same thing here with your models including Cypriots (or any other non-Jewish Eastern Mediterranean population for that matter).
You might - but if several models show similar results - I've used modern and ancient populations with similar components and got similar results. I've actually tested Italian (South and North) in these models, as I've shown, which yield bogus results.

I've actually looked and found that Ashkenazi Jews and Cypriots share similar Y-DNA and mtDNA subclades. For example E-M34. Frequency in Greek Cypriots: 10%. Frequency in Ashkenazi Jews: 10%.
Also I've found here on this forum a thread about additional subclade shared by Cypriots and Ashkenazi Jews, which you, Agamemnon, is a participant:
Ashkenazi 'large group' Y6923 turns up in Cyprus (https://anthrogenica.com/showthread.php?11035-Ashkenazi-large-group-Y6923-turns-up-in-Cyprus)

Erikl86
06-07-2018, 01:59 PM
use tuscan instead of bergamo aschenazis show only 3-4% north italian in living dna
they show much higher south italy and tuscany
regards
adam

p.s
it is cool to see the southern french in your Sephardi run again it is inline with what i see in full sefhardi results gedmatch , gencove , and dna land

Just tested it:

Full nMonte:

[1] "distance%=0.1911 / distance=0.001911"

Ashkenazi_Jew:average

Cypriot:Cyprus22AJ19 52.65
Cypriot:Cyprus21AJ19 27.75
Ukrainian_West:UkrLv240 12.20
French_South:SouthFrench3947 6.35
Italian_Tuscan:NA20509 0.95
Cypriot:Cyprus24AJ19 0.05
Italian_Tuscan:NA20799 0.05
Moroccan:MCA37 0.00
Tunisian:Tunisian20B4 0.00
Turkish:Turkey1 0.00
Turkish:Turkey2 0.00
Turkish:Turkey3 0.00
Turkish:Turkey4 0.00
Turkish:Turkey5 0.00
French_South:SouthFrench3326 0.00
French_South:SouthFrench4018 0.00
Ukrainian_West:UkrLv228 0.00
Ukrainian_West:UkrLv237 0.00
French_East:French24178 0.00
French_East:French24433 0.00
French_East:French24434 0.00
Italian_Tuscan:NA20761 0.00

And restricted nMonte:

[1] "distance%=0.2056 / distance=0.002056"

Ashkenazi_Jew:average

Cypriot:Cyprus21AJ19 46.2
Cypriot:Cyprus22AJ19 32.1
French_South:SouthFrench3947 13.9
Ukrainian_West:UkrLv240 7.8


Seems to have almost no influence, just slight decrease in Levantine-Aegean admixture (of about 5%, perhaps a slight overlap with Aegean admixture as a result of S. Italian admixture) and a little shift towards Northern Europeans (especially southern French).


EDIT: Seems I was right regarding overlap, check out what happens when I add Samaritan:

[1] "distance%=0.1936 / distance=0.001936"

Ashkenazi_Jew:average

Samaritan:GSM537032 25.15
Cypriot:Cyprus21AJ19 23.05
Cypriot:Cyprus22AJ19 20.00
Ukrainian_West:UkrLv240 14.70
Italian_Tuscan:NA20509 8.55
French_South:SouthFrench3947 5.20
French_East:French24434 3.35


Also, again, North/Eastern European admixture got too high (+22%). Samaritan got "only" 25% as opposed to +30% I got when not using Tuscan samples.
This would be expected for the overlap of Samaritans with Cypriots regarding Levantine admixture, but, also for Tuscans which should have a little bit higher Levantine admixture than Bergamo because they are closer to S. Italians and probably received more admixture.

Targum
06-07-2018, 02:15 PM
Yemenites, many of whom have preserved genealogy well, are descended mostly from Jews from Israel , not Babylonia. In fact, their "Teymonith" Hebrew is the same pronunciation as "Ashkenazis", just without the phonological deterioration caused by the Ashkenazi multiple language shifts (Hebrew-Aramaic-Greek-Romance-Yiddish-Slavonic host language), but once an Ashkenazi gets used to it , and vice versa (a story for another time)it is essentially the same pronunciation( and distinct from the Babylonian/origin Sefaradi pronunciation. They like Mizrahhim, lack the So Euro , and instead have admix of So Arabian from the Himyarite royal conversion to Judaism in the pre-Islamic period.
They actively participated in the second Jewish Revolt of Bar Kokhva as financial contributors (the fundraiser being none other than the great Rabbi 'Aqiva the Tanna). Some wealthy 1st century Yemenites are buried in the Jewish catacombs at Beit Shearim, a major historical attraction on the Haifa/Tiberius road in Israel.

Agamemnon
06-07-2018, 02:24 PM
Yemenite Jews seem to be mostly Himyarite in origin, with additional Judean ancestry (probably via some of the Jewish tribes of Arabia, which were of Kohanic origin by the way). This is why they look like a pristinely Arabian population, unlike Mizrahi Jews who owe most of their ancestry to a Mesopotamian population (with a more substantial Levantine component).

And Targum is right, the Yemenite pronounciation of Hebrew is extremely similar to the Ashkenazi pronounciation system (especially as far as the spirantisation of non-stop consonants and the pronounciation of the vowels are of concern, as well as the generalisation of penultimate stress), I would actually argue that the loss of the gutturals in Ashkenazi Hebrew probably did not come about via contact-induced change with European languages, rather they were probably lost by the time the Romans destroyed Jerusalem.

kingjohn
06-07-2018, 02:30 PM
Yemenites, many of whom have preserved genealogy well, are descended mostly from Jews from Israel , not Babylonia. In fact, their "Teymonith" Hebrew is the same pronunciation as "Ashkenazis", just without the phonological deterioration caused by the Ashkenazi multiple language shifts (Hebrew-Aramaic-Greek-Romance-Yiddish-Slavonic host language), but once an Ashkenazi gets used to it , and vice versa (a story for another time)it is essentially the same pronunciation( and distinct from the Babylonian/origin Sefaradi pronunciation. They like Mizrahhim, lack the So Euro , and instead have admit if Himyarite So Arabian from the Himyarite royal conversion to Judaism in the ore/Islamic period.

they yemenite jews have arabian admixture
and mizrachi from iraq have mesopotamian admixture :)

Erikl86
06-07-2018, 02:42 PM
Using no Cypriots and only Samaritans, and Bergamo Italians as "south" Europeans (Italians), I get bogus results (but short distance, thanks to Levantine and probably correct North/East components):

[1] "distance%=0.2138 / distance=0.002138"

Ashkenazi_Jew:average

Samaritan:GSM537032 60.40
French_East:French24434 19.20
Ukrainian_West:UkrLv240 15.95
Italian_Bergamo:HGDP01174 4.45

Let's try Tuscan and Samaritan, no Cypriot:

[1] "distance%=0.2185 / distance=0.002185"

Ashkenazi_Jew:average

Samaritan:GSM537032 58.4
Ukrainian_West:UkrLv240 18.4
French_East:French24434 13.3
Italian_Tuscan:NA20761 9.9

Again, bogus results. Italian vs Levantine is way too low, and North/Eastern component is way, way too high. But you can see the overlap I was talking about - Tuscans lower the percentage of Samaritans a bit because of small Levantine admixture, probably originating from S. Italians migrating north.

Erikl86
06-07-2018, 02:53 PM
All so called ancient genealogy documentations of Jews lead to Israel :).

Yemenite tradition itself claims the community exist before the destruction of the 1st Temple, which I doubt.

Also, another legend states that when Ezra commanded the Jews to return to Jerusalem they disobeyed, whereupon he pronounced a ban upon them. According to this legend, as a punishment for this hasty action, Ezra was denied burial in Israel. As a result of this local tradition, which can not be validated historically, it is said that no Jew of Yemen gives the name of Ezra to a child.

So if judging simply by tradition etc., Yemenite community should have existed since the Babylonian captivity. I know the first archaeological records put Jews in Yemen since the 2nd century BC. So if their main origin is Judean founding men which arrived during the Babylonian captivity to the area, or just before Hellenistic period, maybe during the Persian period, they would have no significant Aegean DNA.

As for Yemenite Hebrew, it looks similar to Ashkenazic Hebrew, but it is completely different. Also, liturgical Hebrew pronunciation might simply be a matter of preference of the community to follow Babylonian jurisdiction or Tiberian one. After all, other than Ethiopian Jews, most Jewish communities have kept some connection with important Jewish centers throughout the ages.

Agamemnon
06-07-2018, 03:03 PM
There is no indication that Yemenite Jewry dates back to the Babylonian exile, the arrival of the Jews in Tayma and Yathrib dates back to the late classical period for example while the emergence of Yemenite Judaism coincides with the adoption of Judaism by the Himyarites.

The explanation you put forth sounds more like special pleading quite frankly, the most parsimonious one is that the Judeans were a pristinely Levantine population which had little to no Aegean admixture. There is strictly no need to invoke some convoluted migration from Mesopotamia to Arabia.

The different pronounciation systems of Hebrew are basically eroded forms of the Tiberian one, which itself has its roots in Mishnaic pronounciation. The Tiberian pronounciation is the standard when it comes to establishing the pronounciation of the vowels. On the other hand, a comparison with Samaritan Hebrew is also warranted since there are similarities to be found with Ashkenazi and Sephardic Hebrew pronounciation.

Targum
06-07-2018, 03:10 PM
Yemenites name children 'Ezra all the time and always have. We have 'Ezras in our extended family named after prior 'Ezra's and I found it to be a name more common among Teymanim.

Erikl86
06-07-2018, 03:19 PM
So, I've checked Yemenite Jews, and indeed, they have no Cypriot, nor Assyrian admixture. They are extremely close to Saudis - closer than Western Jews are to Sicilian or S. Italians.

I've ran Saudi with Samaritan:

[1] "distance%=0.98 / distance=0.0098"

Yemenite_Jew:average

Saudi:saudi1403 76.0
Saudi:saudi1424 23.9

Zero results for Samaritans. Might be because Saudis are also have some small Levantine admixture, or Samaritans might got some south Arabian admixture. I don't know - this is indeed strange. Or, it's just that Yemenite Jews are extremely mixed with Arabs.

Anyhow, I did try to add ancient Aegean samples, and this is what I got:

[1] "distance%=0.8996 / distance=0.008996"

Yemenite_Jew:average

Saudi:saudi1403 48.6
Saudi:saudi1424 47.5
Minoan_Lasithi:I9005 3.9

Minoan appeared. Could it be from Philistine admixture with Judeans? This is quite small - about 4%. Or maybe, if indeed Yemenite Jewish community founders arrived in the 2nd century BC, not enough time for significant admixture with Hellenistic Greek?

I've tried Iraqi Jews:

[1] "distance%=0.1199 / distance=0.001199"

Iraqi_Jew:average

Assyrian:GS000013749 75.2
Samaritan:GSM537032 24.8
Minoan_Lasithi:I0070 0.0
Minoan_Lasithi:I0071 0.0
Minoan_Lasithi:I0073 0.0
Minoan_Lasithi:I0074 0.0
Minoan_Lasithi:I9005 0.0
Minoan_Odigitria:I9130 0.0
Minoan_Odigitria:I9131 0.0
Mycenaean:I9006 0.0
Mycenaean:I9010 0.0
Mycenaean:I9041 0.0
Assyrian:GS000013750 0.0
Assyrian:GS000013751 0.0

Zero admixture with Ancient aegeans.

With only modern populations:

[1] "distance%=0.1199 / distance=0.001199"

Iraqi_Jew:average

Assyrian:GS000013749 75.2
Samaritan:GSM537032 24.8
Assyrian:GS000013750 0.0
Assyrian:GS000013751 0.0
Cypriot:Cyprus21AJ19 0.0
Cypriot:Cyprus22AJ19 0.0
Cypriot:Cyprus24AJ19 0.0


Zero admixture with Cypriots as well.

Erikl86
06-07-2018, 03:21 PM
Yemenites name children 'Ezra all the time and always have. We have 'Ezras in our extended family named after prior 'Ezra's and I found it to be a name more common among Teymanim.

I've heard this story before, and wikipedia also include this legend. I wonder where it comes from.

Agamemnon
06-07-2018, 03:22 PM
Some wealthy 1st century Yemenites are buried in the Jewish catacombs at Beit Shearim, a major historical attraction on the Haifa/Tiberius road in Israel.

The bilingual Sabaic-Aramaic inscriptions from Beit She'arim are quite fascinating as well. Here's an example:

https://www.persee.fr/renderIllustration/crai_0065-0536_2004_num_148_2_T1_0841_0000_1.png

kingjohn
06-07-2018, 03:25 PM
So, I've checked Yemenite Jews, and indeed, they have no Cypriot, nor Assyrian admixture. They are extremely close to Saudis - closer than Western Jews are to Sicilian or S. Italians.

I've ran Saudi with Samaritan:

[1] "distance%=0.98 / distance=0.0098"

Yemenite_Jew:average

Saudi:saudi1403 76.0
Saudi:saudi1424 23.9

Zero results for Samaritans. Might be because Saudis are also have some small Levantine admixture, or Samaritans might got some south Arabian admixture. I don't know - this is indeed strange. Or, it's just that Yemenite Jews are extremely mixed with Arabs.

Anyhow, I did try to add ancient Aegean samples, and this is what I got:

[1] "distance%=0.8996 / distance=0.008996"

Yemenite_Jew:average

Saudi:saudi1403 48.6
Saudi:saudi1424 47.5
Minoan_Lasithi:I9005 3.9

Minoan appeared. Could it be from Philistine admixture with Judeans? This is quite small - about 4%. Or maybe, if indeed Yemenite Jewish community founders arrived in the 2nd century BC, not enough time for significant admixture with Hellenistic Greek?

I've tried Iraqi Jews:

[1] "distance%=0.1199 / distance=0.001199"

Iraqi_Jew:average

Assyrian:GS000013749 75.2
Samaritan:GSM537032 24.8
Minoan_Lasithi:I0070 0.0
Minoan_Lasithi:I0071 0.0
Minoan_Lasithi:I0073 0.0
Minoan_Lasithi:I0074 0.0
Minoan_Lasithi:I9005 0.0
Minoan_Odigitria:I9130 0.0
Minoan_Odigitria:I9131 0.0
Mycenaean:I9006 0.0
Mycenaean:I9010 0.0
Mycenaean:I9041 0.0
Assyrian:GS000013750 0.0
Assyrian:GS000013751 0.0

Zero admixture with Ancient aegeans.

With only modern populations:

[1] "distance%=0.1199 / distance=0.001199"

Iraqi_Jew:average

Assyrian:GS000013749 75.2
Samaritan:GSM537032 24.8
Assyrian:GS000013750 0.0
Assyrian:GS000013751 0.0
Cypriot:Cyprus21AJ19 0.0
Cypriot:Cyprus22AJ19 0.0
Cypriot:Cyprus24AJ19 0.0


Zero admixture with Cypriots as well.

kudos you nail it with the Assyrian stuff :beerchug:
kind regards
adam

p.s
if i send you my ftdna raw data can you run me ?

Erikl86
06-07-2018, 03:26 PM
There is no indication that Yemenite Jewry dates back to the Babylonian exile, the arrival of the Jews in Tayma and Yathrib dates back to the late classical period for example while the emergence of Yemenite Judaism coincides with the adoption of Judaism by the Himyarites.

The explanation you put forth sounds more like special pleading quite frankly, the most parsimonious one is that the Judeans were a pristinely Levantine population which had little to no Aegean admixture. There is strictly no need to invoke some convoluted migration from Mesopotamia to Arabia.


Check out my results. Iraqi Jews get 0 aegean admixture, Yemenite Jews get very small ancient aegean admixture. Could be either from Philistines (which, if they indeed came from Crete, would probably be closer to Minoans in their Aegean admixture, as they've arrived the Levant by the 12th century BC, not that long ("only" 200 years) after the Minoan civilization disappeared) admixture in Judean Jews, which occurred mainly after the Babylonian captivity, or perhaps from some Hellenistic Greeks which had closer genetic links to Minoans for some reason (or, again, nMonte simply has some difficulties distinguishing between the two, as they are very close).

Targum
06-07-2018, 03:35 PM
The bilingual Sabaic-Aramaic inscriptions from Beit She'arim are quite fascinating as well. Here's an example:

https://www.persee.fr/renderIllustration/crai_0065-0536_2004_num_148_2_T1_0841_0000_1.png


Wow! Thanks @Agamemnon. It looks like "הדה קבורתה לאה ברת יהודה...״". "hadah qevurtah Leah barat Yehudah.." Here is interred Leah the daughter of Yehudah... any contemporary Hebrew reader can read the script clearly

Agamemnon
06-07-2018, 03:52 PM
Check out my results. Iraqi Jews get 0 aegean admixture, Yemenite Jews get very small ancient aegean admixture. Could be either from Philistines (which, if they indeed came from Crete, would probably be closer to Minoans in their Aegean admixture, as they've arrived the Levant by the 12th century BC, not that long ("only" 200 years) after the Minoan civilization disappeared) admixture in Judean Jews, which occurred mainly after the Babylonian captivity, or perhaps from some Hellenistic Greeks which had closer genetic links to Minoans for some reason (or, again, nMonte simply has some difficulties distinguishing between the two, as they are very close).

This is a fit I obtained last february.

Scaled (w/o pen=0):

[1] "distance%=2.121"

Yemenite_Jew

Yemenite_Mahra,78
Samaritan,22

^^That looks a lot more plausible than your fit where they are basically 96% Saudi + 4% Minoan.


Wow! Thanks @Agamemnon. It looks like "הדה קבורתה לאה בת יהודה...״". Here is interred Leah the daughter of Yehudah... any contemporary Hebrew reader can read the script clearly

Yes, the inscription reads:

הדה קבורתה דלאה ברת יהודה נשמתה לחיי עולם ותנוח ותעמוד לגורל חיים לקץ הימין אמן ואמן שלום

Which is best translated as:

This is the tomb of Leah, daughter of Yehudah. May her soul rest for eternal life, she shall rest and stand for resurrection at the end of days. Amen and amen. Shalom.

The Sabaic part is slightly different, the text reads:

qbwrt lʾh bt ywdh lnḥnhw rḥmnn ʾmn s¹lwm

Which is best translated as:

Tomb of Leah, daughter of Yawdah (ywdh), May Raḥmānān (rḥmnn) grant her rest. Amen. Shalom.

Targum
06-07-2018, 04:00 PM
This is a fit I obtained last february.

Scaled (w/o pen=0):

[1] "distance%=2.121"



Yemenite_Jew

Yemenite_Mahra,78
Samaritan,22

^^That looks a lot more plausible than your fit where they are basically 96% Saudi + 4% Minoan.



Yes, the inscription reads:

הדה קבורתה דלאה ברת יהודה נשמתה לחיי עולם ותנוח ותעמוד לגורל חיים לקץ הימין אמן ואמן שלום

Which is best translated as:

This is the tomb of Leah, daughter of Yehudah. May her soul rest for eternal life, she shall rest and stand for resurrection at the end of days. Amen and amen. Shalom.

The Sabaic part is slightly different, the text reads:

qbwrt lʾh bt ywdh lnḥnhw rḥmnn ʾmn s¹lwm

Which is best translated as:

Tomb of Leah, daughter of Yawdah (ywdh), May Raḥmānān (rḥmnn) grant her rest. Amen. Shalom.

Which is virtually identical to both contemporary Jewish headstone inscriptions, as well as versions of the "א-ל מלא רחמים"
"E-l male rahhamim" memorial prayer.

Erikl86
06-07-2018, 04:03 PM
[1] "distance%=2.121"

Yemenite_Jew

Yemenite_Mahra,78
Samaritan,22

If only I had non-Jewish Yemenite Basal-rich K7 stats :-( I'd use them. Unfortunately the closest I have are the Saudis. Being that Saudi Arabians were basically Bedouins which have been travelling from the Syrian desert all the way to Hejaz, it might be that these Saudis have some Levantine admixture, enough to overlap with Samaritans.

I'd try to run on the K36oracle calculator, where I have Yemenites, but it's less accurate for ancient populations than Basal-rich K7.

Eihwaz
06-07-2018, 04:06 PM
If only I had non-Jewish Yemenite Basal-rich K7 stats :-( I'd use them. Unfortunately the closest I have are the Saudis. Being that Saudi Arabians were basically Bedouins which have been travelling from the Syrian desert all the way to Hejaz, it might be that these Saudis have some Levantine admixture, enough to overlap with Samaritans.

I'd try to run on the K36oracle calculator, where I have Yemenites, but it's less accurate for ancient populations than Basal-rich K7.

I'd try running these populations in G25 if I were you... :P

Erikl86
06-07-2018, 06:59 PM
I'd try running these populations in G25 if I were you... :P

No luck with Eurogenes K36 Oracle either... the reference spreadsheet has Yemenites, but no Yemenite Jews, and while I have Iraqi, Iranian, Ashkenazi etc., GEDmatch numbers, I have no Yemenite Jews. The one I used is an average from the reference spreadsheet of Eurogenes Basal-rich K7. If someone has access to Eurogenes K36 results for Yemenite Jew, or a kit number, then I'll run it along the ancient aegean kits which I have.

K33
06-07-2018, 08:34 PM
Yemeni Jews IIRC have higher Basal Eurasian/Natufian ancestry and lower East African ancestry than their Muslim neighbors, which I'd say implies more pronounced continuity from the Mesolithic (even if the y-dna shows some founder effects from Israel) relative to Muslims. Babylonian Jews would have been more CHG-shifted.

Also @Agememmnon, I'm not sure I agree with this:


Jewish proselytism was restricted to very specific episodes, such as the expansion of the Hasmonean kingdom (which brought about the forced conversions of the Idumeans, the Itureans and many Samaritans), the conversion of the Himyarites to Judaism or the conversion of the Khazars to Judaism. Such episodes were the exception, not the rule. As a rule of thumb, Judaism was (and still is) much more akin to the religions of the ancient world since the entry was primarily biological.I find Oswald Spengler's identification of 3 distinct phases of Judaism to be convincing; the first is Judaism as an old paganeque tribal religion of the Israelite people; the second phase begins around the 6th/7th century BC with the great prophets like Isaiah, which eventually turned Judaism into a genuinely henotheistic "community of believers" detached from locality (and opening up the prospect of conversions and proselytization), then finally by the early middle ages this organic growth had ended and Judaism once again became more "tribal" and insular.

You already mentioned the Himyarite Kingdom and Khazaria during this period; we know that the Jewish population of Egypt and Iran had grown quite significant by the late Seleucid period, Adiabene in Iraq which became a Jewish state during the early Roman Empire, as Erikl noted the conversion of Greeks to Judaism is documented by various sources; there is the case of other Arabian Jewish converts like Samuel of Al-Ablaq (https://en.wikipedia.org/wiki/Samaw%27al_ibn_%27Adiya), and its likely that at least some of the Beta Israel Jews were converted at this time (and not earlier, during the Solomonian period). And don't forget the case of the Kaifeng Jews; it seems that like the Manicheans and later the Muslims, Judaism made more inroads into Central Asia than some may think...

Obviously, whether these groups of proselytized Jews contributed to modern Jewry (and if so to what extent) is still up for debate. But I think it's tough to believe that all these Jewish communities springing up around the world during this era represents the demic diffusion of Jews from Israel...

IIRC the rule of matrilineal descent only dates to the early middle ages for Ashkenazim, right?

Agamemnon
06-07-2018, 09:04 PM
Also @Agememmnon, I'm not sure I agree with this:

I find Oswald Spengler's identification of 3 distinct phases of Judaism to be convincing; the first is Judaism as an old paganeque tribal religion of the Israelite people; the second phase begins around the 6th/7th century BC with the great prophets like Isaiah, which eventually turned Judaism into a genuinely henotheistic "community of believers" detached from locality (and opening up the prospect of conversions and proselytization), then finally by the early middle ages this organic growth had ended and Judaism once again became more "tribal" and insular.

You already mentioned the Himyarite Kingdom and Khazaria during this period; we know that the Jewish population of Egypt and Iran had grown quite significant by the late Seleucid period, Adiabene in Iraq which became a Jewish state during the early Roman Empire, as Erikl noted the conversion of Greeks to Judaism is documented by various sources; there is the case of other Arabian Jewish converts like Samuel of Al-Ablaq (https://en.wikipedia.org/wiki/Samaw%27al_ibn_%27Adiya), and its likely that at least some of the Beta Israel Jews were converted at this time (and not earlier, during the Solomonian period). And don't forget the case of the Kaifeng Jews; it seems that like the Manicheans and later the Muslims, Judaism made more inroads into Central Asia than some may think...

Obviously, whether these groups of proselytized Jews contributed to modern Jewry (and if so to what extent) is still up for debate. But it's of course impossible to posit that all these Jewish communities springing up around the world during this era represents the demic diffusion of Jews from Israel...

IIRC the rule of matrilineal descent only dates to the early middle ages for Ashkenazim, right?

While Spengler offered some very interesting insights on the rise and fall of civilisations, his characterisation of Jewish history is somewhat detached from what the evidence now shows. The Israelites originally were every bit as pagan as their neighbours, in fact their very name suggests that they worshipped El (who was the head of the Canaanite pantheon), accordingly Yahweh is a deity that seems to have arrived from the south (probably with the famous "Shasu of Yhw"). The appearance of monolatry was gradual, during the first phases of the Iron Age (Iron Age I Israelite religion) Yahweh was assimilated to both El and Ba'al, the existence of other gods wasn't denied. During the last phases of the Iron Age II we start to see a form of henotheism very much akin to the form of henotheism practiced by the Babylonians and the Assyrians, in the Elephantine papyri we learn that Yahweh still had a consort ('Anat-Yahu). After the Babylonian exile, the existence of other gods is completely denied, and Yahweh's consort has been either erased or assimilated to the "Shekhinah" (God's presence). With the rise of the Hasmonean kingdom, Judaism takes a recognisable form, with the destruction of the second temple it became a mobile religion and power shifted from the priests to the rabbis.

As I said earlier, it is a mistake to think of Judaism as a faith akin to Christianity or Islam. Save a few exceptions, it always was "tribal" as the entry was primarily biological, most of Jewish population centers you've cited have more to do with Jews taking local partners than with mass conversions in Egypt, Mesopotamia and Iran. The only way to "convert" back then was by marrying a Jew. We have to wait until the Tanna'im to see 1) the emergence of the matrilineal principle and 2) a real conversion process (which didn't involve invading an area and forcibly converting the population). Even then, conversion probably had nothing to do with what Jewish conversions are like today. I am of course familiar with Samuel Ibn Adiya, if anything his case just serves to illustrate my point (you'll notice that some have traced his "nasb" back to the Kohanim of Jerusalem, which is in line with the fact that most Arabian Jewish tribes were of Kohanic descent), I could also point to Josephus' antiquities to highlight how "conversion" was akin to intermarriage prior to the Tanna'im. Also, Ethiopian Jews have no real Mosaic tradition, they do not even use Hebrew as a liturgical language, and indeed they do not seem to have actual Jewish ancestry.

Like I said, the matrilineal principle dates back to the Tanna'im, so around the first two centuries CE (probably after the destruction of the second temple though). Its original purpose was to extend Jewishness to the offspring of non-Jewish fathers, as the law originally was patrilineal (similar to the Samaritan halakhah). In time, it came to be understood as a rejection of patrilineal descent, it was seriously applied starting from the early Middle Ages (throughout the Jewish world, though there were some exceptions, such as the Karaites and more isolated Jewish communities in the Caucasus).

Targum
06-07-2018, 09:35 PM
No @K33; matrilineal descent determining Jewishness( as opposed to the patrilineal tribal affiliation) is attested to in the Oral Torah already in the Mishnah (late Roman)period . It was written down then and reflects an older oral tradition, as what became the Mishnayot, had been memorized, passed from Master (Rav) to Disciple (Talmid) for generations. Tradition teaches that the Oral Torah was given at Sinai, but in any case;it is older. The writing down of the Mishnah in the time if, and at the direction of, Rabbi Yehudah HaNasi due to the dangerous Roman persecutions and general social decline in Israel, reflected much older memorized oral tradition.Even today we ( I, son, grandsons etc) learn Mishnah with a tune, like a maqam.
As to the assumption in another part of the question, I humbly wish to correct your mistaken information . Local customs and interpretations vary whether one is Yemenite, Sefaradi, Ashkenazi, or Mizrahhi, but core Halakhah or Jewish Law, does not.Matrilineal descent is a core halakhah held by all observant Jews. What varied are sometimes applications and customs. For example the pe'ot or " corners of the beard" cannot be cut , but different communities have set the boundaries from, shaving the rest of the face with an electric razor all the way to never shaving, and I can give many such examples.

Agamemnon
06-07-2018, 09:59 PM
We know from what Josephus, Philo of Alexandria and other ancient writers have written that the matrilineal law was virtually unheard of during the 1st century CE. There's also the fact that no matter which model we're dealing with here, as far as the ethnogenesis of most Jewish groups (Western, Mizrahi, Yemenite) is of concern it always involves Jewish men taking local brides, hence the wealth of Levantine Y-chromosomal lineages and the paucity of Levantine mitochondrial lineages in present-day Jews (you're actually a pretty good example of this yourself), which goes on to underline the fact that the matrilineal principle was not always strictly applied. Some Jewish communities, such as the Juhurim, even kept patrilineal descent as the main standard for Jewishness. The Karaite patrilineal law might be a later development, though it is interesting that in many ways their doctrine resembles that of the Sadducees.

Targum
06-07-2018, 10:16 PM
We know from what Josephus, Philo of Alexandria and other ancient writers have written that the matrilineal law was virtually unheard of during the 1st century CE. There's also the fact that no matter which model we're dealing with here, as far as the ethnogenesis of most Jewish groups (Western, Mizrahi, Yemenite) is of concern it always involves Jewish men taking local brides, hence the wealth of Levantine Y-chromosomal lineages and the paucity of Levantine mitochondrial lineages in present-day Jews (you're actually a pretty good example of this yourself), which goes on to underline the fact that the matrilineal principle was not always strictly applied. Some Jewish communities, such as the Juhurim, even kept patrilineal descent as the main standard for Jewishness. The Karaite patrilineal law might be a later development, though it is interesting that in many ways their doctrine resembles that of the Sadducees.

ahhi hayaqar; You do not "Pasqen' fromYosef ben Matityahu; but from the Gemara quoting a Mishnah you have proof of the historicity of Matrilneal descent in Tractate Qiddushin (betrothal Contracts). All the Italian, Persian etc women were converts, and conversion was less draconian than today. the book of 'Ezra and Nehemiah shows the attitude from hunreds of years earlier. The Talmud presumes paternity, pre-DNA, was a shaky concept, especially in promiscuous cultures as the Torah perceives Egypt and Canaan.

Erikl86
06-07-2018, 10:37 PM
No luck with Eurogenes K36 Oracle either... the reference spreadsheet has Yemenites, but no Yemenite Jews, and while I have Iraqi, Iranian, Ashkenazi etc., GEDmatch numbers, I have no Yemenite Jews. The one I used is an average from the reference spreadsheet of Eurogenes Basal-rich K7. If someone has access to Eurogenes K36 results for Yemenite Jew, or a kit number, then I'll run it along the ancient aegean kits which I have.

Found one, checked if he's actually Yemenite Jew on GEDmatch:


K13 Oracle ref data revised 21 Nov 2013

Admix Results (sorted):

# Population Percent
1 East_Med 52.82
2 Red_Sea 28.50
3 Northeast_African 6.68
4 West_Asian 6.49
5 West_Med 5.29


Finished reading population data. 204 populations found.
13 components mode.

--------------------------------

Least-squares method.

Using 1 population approximation:
1 Yemenite_Jewish @ 1.310451
2 Saudi @ 12.439169
3 Bedouin @ 19.429731
4 Egyptian @ 19.674311
5 Palestinian @ 19.848322
6 Jordanian @ 22.809244
7 Samaritan @ 23.612978
8 Lebanese_Christian @ 25.151546
9 Lebanese_Druze @ 26.379465
10 Syrian @ 27.870626
11 Libyan_Jewish @ 29.078968
12 Tunisian_Jewish @ 29.432554
13 Lebanese_Muslim @ 29.770828
14 Cyprian @ 32.226704
15 Kurdish_Jewish @ 33.102966
16 Iranian_Jewish @ 33.138550
17 Sephardic_Jewish @ 34.233524
18 Algerian_Jewish @ 34.687027
19 Italian_Jewish @ 34.834064
20 Tunisian @ 34.846348

Using 2 populations approximation:
1 50% Yemenite_Jewish +50% Yemenite_Jewish @ 1.310451


Using 3 populations approximation:
1 50% Yemenite_Jewish +25% Yemenite_Jewish +25% Yemenite_Jewish @ 1.310451


Using 4 populations approximation:
++++++++++++++++++++++++++++++++++++++++++++++++++ +++++++++++++++++++++
1 Yemenite_Jewish + Yemenite_Jewish + Yemenite_Jewish + Yemenite_Jewish @ 1.310451
2 Saudi + Yemenite_Jewish + Yemenite_Jewish + Yemenite_Jewish @ 3.688180
3 Egyptian + Yemenite_Jewish + Yemenite_Jewish + Yemenite_Jewish @ 5.132768
4 Bedouin + Yemenite_Jewish + Yemenite_Jewish + Yemenite_Jewish @ 5.282885
5 Palestinian + Yemenite_Jewish + Yemenite_Jewish + Yemenite_Jewish @ 5.334513
6 Jordanian + Yemenite_Jewish + Yemenite_Jewish + Yemenite_Jewish @ 5.982632
7 Samaritan + Yemenite_Jewish + Yemenite_Jewish + Yemenite_Jewish @ 6.267070
8 Saudi + Saudi + Yemenite_Jewish + Yemenite_Jewish @ 6.557342
9 Lebanese_Christian + Yemenite_Jewish + Yemenite_Jewish + Yemenite_Jewish @ 6.623398
10 Egyptian + Saudi + Yemenite_Jewish + Yemenite_Jewish @ 6.807082
11 Palestinian + Saudi + Yemenite_Jewish + Yemenite_Jewish @ 6.869149
12 Lebanese_Druze + Yemenite_Jewish + Yemenite_Jewish + Yemenite_Jewish @ 6.916992
13 Syrian + Yemenite_Jewish + Yemenite_Jewish + Yemenite_Jewish @ 7.370462
14 Samaritan + Saudi + Yemenite_Jewish + Yemenite_Jewish @ 7.456209
15 Bedouin + Saudi + Yemenite_Jewish + Yemenite_Jewish @ 7.462330
16 Jordanian + Saudi + Yemenite_Jewish + Yemenite_Jewish @ 7.566741
17 Lebanese_Christian + Saudi + Yemenite_Jewish + Yemenite_Jewish @ 7.648847
18 Lebanese_Druze + Saudi + Yemenite_Jewish + Yemenite_Jewish @ 7.662932
19 Libyan_Jewish + Yemenite_Jewish + Yemenite_Jewish + Yemenite_Jewish @ 7.669492
20 Tunisian_Jewish + Yemenite_Jewish + Yemenite_Jewish + Yemenite_Jewish @ 7.777946

This is his Eurogenes K36 Oracle admixture:

1. When I compared to Hadramut Yemenites, another sample of Yemenite from Aden, and a Samaritan, I got huge distance:

[1] "distance%=27.5486 / distance=0.275486"

Yemenite_Jew

Yemen_Aden 51.1
Samaritan 48.9


2. Tried to introduce Mehri Yemenites, and got the following results:

[1] "distance%=2.6941 / distance=0.026941"

Yemenite_Jew

Yemen_Mehri1 49.20
Yemen_Mehri2 42.55
Yemen_Aden 8.25
Samaritan 0.00
Hadramut_Yemen 0.00

Pretty good results for Eurogenes K36, but - zero for Samaritans.


Now let's check with BA_Levant instead of Samaritans, because perhaps they are a little bit too drifted as a result of their isolation, or maybe Mehri, which seem to be a little more "Levantine", simply overlaps them. Maybe we'll get something better:

[1] "distance%=2.8958"

Yemenite_Jew

Yemen_Mehri1,52
Yemen_Mehri2,42.6
Yemen_Aden,5.4

Still, no Levantine shows up. Let's check now with Levant_N + CHL_Iranian:

[1] "distance%=3.0033"

Yemenite_Jew

Yemen_Mehri1,54.4
Yemen_Mehri2,40.2
Yemen_Aden,3.4
Hadramut_Yemen,1.6
N_Levant_AinGhazal5,0.4

Ok, so let's try again, this time removing the Mehri, which perhaps overlap too much, and using only the Aden one (Hadramut Yemenis seem to be pretty far off Yemeni Jews), but using N_Levant / CHL_Iran to model Levantines:

[1] "distance%=26.493"

Yemenite_Jew

N_Levant_AinGhazal1,59
Yemen_Aden,31.8
Hadramut_Yemen,9.2

Again, pretty far off and inaccurate.

Now let's try the BA_Sidon/Levant, again without Mehri:

[1] "distance%=25.1726"

Yemenite_Jew

Yemen_Aden,37.4
BA_Levant_I0867,33.4
BA_Sydon,29.2

Again, bogus results.

It seems to me that Mehri people are partially Levantine, which is why they give such good results, but completely overlap with any other Levantine population. They also probably represent a closer model to the non-Jewish Yemenis which converted to Judaism when they joined the Yemenite Jewish community.



Now, let's try Cypriot, with Mehri again:

distance%=2.9411"

Yemenite_Jew

Yemen_Mehri1,53.4
Yemen_Mehri2,40.8
Yemen_Aden,3.4
Hadramut_Yemen,2.4

Nothing.

Now, let's try to add some other Greek Islanders:

[1] "distance%=2.544"

Yemenite_Jew

Yemen_Mehri1,48
Yemen_Mehri2,34.6
Yemen_North,14.2
Yemen_Aden,2.8
Hadramut_Yemen,0.4

My opinion, is that Mehri Yemenis, the only ones which produce any close results, are also mixed with Levantines, which overlaps any attempt to use Levantines in the admixture. Problem is, if I take the Mehri out, the distance is just too big.
It's either that, or that Yemenite Jews are mostly descendant from Yemeni converts. I wish I had more Yemenite Jewish samples. I'm afraid it maybe not diverse enough, despite the fact that it got such a short distance from Yemenite Jews in it's Oracle.

I'm still not convinced exactly when Yemenite Jews have left Judea. Again, if it was before Hellenistic period, it might very well be that they will not have substantial Aegean admixture. Right now, at least that Yemenite Jews I've found, seem to have no distinguishable Levantine Yemenite admixture.

I'll try the One-to-Many on GEDmatch to try and locate more Yemeni Jews, and try to create an average. By far, even Iraqi Jews are not that non-Levantine (again, this might be for two reasons: this specific Yemeni Jew is descendant of converts, and Mehri Yemenis are already heavily mixed with Levantines).

Sikeliot
06-08-2018, 12:24 AM
There are a number of misunderstandings in this thread and as such I have clarified most of them below.

Western Jews descend from a population of Jews who lived primarily in central and southern Italy and moved northward, not in Greece. They were Greek-speaking because southern Italy as a whole was Greek-speaking, but Greece itself in that era did not have the largest Jewish population in Europe; Italy did. Italkim are probably a good proxy group in Italy for one of the older Jewish groups, who would have migrated north, acquired significant Italic admixture, and then spread into Europe.

It is very difficult to distinguish modern Ashkenazim/Sephardim from southern Italians on an autosomal level today because these parts of Italy have ancient Levantine ancestry that is likely very close to that of the Canaanite populations who became Phoenicians and Jews. However, due to that this admixture does not directly correspond with the parts of southern Italy that received Phoenician, Carthaginian, and Arab/Berber admixture, we should view at least part of this as very ancient admixture that would have been present since very ancient times.

There is a Bronze Age Sicilian GEDmatch result which already has such admixture in significant amounts:



MDLP K23:
# Population Percent
1 European_Early_Farmers 38.26
2 Caucasian 34.77
3 North_African 8.09
4 European_Hunters_Gatherers 6.53
5 Near_East 4.19
6 Australoid 3.51
7 Archaic_Human 1.58
8 African_Pygmy 1.13
9 Tungus-Altaic 1.04


Finished reading population data. 620 populations found.
23 components mode.

--------------------------------

Least-squares method.

Using 1 population approximation:
1 Italian_Abruzzo_ @ 13.761235
2 Italian_Tuscan_ @ 14.247825
3 Italian_Piedmont_ @ 14.386563
4 Sicilian_Trapani_ @ 15.749977
5 Italian_Bergamo_ @ 16.123007
6 Maltese_ @ 16.194090
7 Sicilian_Agrigento_ @ 16.914177
8 Sicilian_Siracusa_ @ 17.301807
9 Sicilian_Center_ @ 17.890909
10 French_Jew_ @ 17.951210
11 Sicilian_West_ @ 18.450773
12 Sicilian_East_ @ 18.476643
13 Spanish_Baleares_IBS_ @ 18.967546
14 Ashkenazi_Jew_ @ 19.462942
15 Ashkenazi_ @ 19.639051
16 Italian_Jew_ @ 19.785477
17 Romanian_Jew_ @ 20.599899
18 Sephardic_Jew_ @ 20.734110
19 Italian_South_ @ 20.863527
20 Moroccan_Jew_ @ 20.915777

Using 2 populations approximation:
1 50% Crimean_Tatar_Coast_ +50% Sardinian_ @ 8.298473


MDLP World22:
# Population Percent
1 Atlantic_Mediterranean_Neolithic 41.36
2 Near_East 26.70
3 West-Asian 22.43
4 North-East-European 4.58
5 North-European-Mesolithic 2.52
6 Austronesian 1.85


Finished reading population data. 276 populations found.
22 components mode.

--------------------------------

Least-squares method.

Using 1 population approximation:
1 Sephardim_derived @ 5.335933
2 Jew_Algeria_derived @ 6.473939
3 Jew_Italia_derived @ 6.937978
4 Jew_Francestrale_derived @ 7.165184
5 Jew_Morocco_derived @ 9.023011
6 Jew_Tunisia_derived @ 9.559836
7 Jew_Libya_derived @ 10.691164
8 Ashkenazim_derived @ 11.281357
9 Sicilian_derived @ 12.441468
10 Cypriot_derived @ 13.258138
11 Greek_Cretan_derived @ 13.432624
12 Jew_Syria_derived @ 14.666184
13 Italian-South_derived @ 15.173383
14 Italian-Center_derived @ 17.368025
15 Greek_East_derived @ 19.032942
16 Jew_Romania_derived @ 19.691177
17 Greek_Center_derived @ 20.506847
18 Lebanese_derived @ 22.316582
19 Costanoan_derived @ 22.689951
20 Druze_derived @ 23.163324

Using 2 populations approximation:
1 50% Jew_Azerbaijan_derived +50% Otzi_derived @ 3.807522

Using 3 populations approximation:
1 50% Jew-Iraqi_derived +25% Otzi_derived +25% Sardinian_derived @ 3.038447

MDLP World:
# Population Percent
1 South_and_West_European 45.36
2 Middle_East 35.67
3 Caucaus_Parsia 13.16
4 North_and_East_European 4.13
5 North_Asian 1.33


Finished reading population data. 257 populations found.
12 components mode.

--------------------------------

Least-squares method.

Using 1 population approximation:
1 Jew_Morocco @ 7.166730
2 Jew_France @ 8.249454
3 Sephardim @ 8.596121
4 Jew_Italia @ 8.938586
5 Jew_Algeria @ 10.401005
6 Jew_Tunisia @ 11.487771
7 Jew_Libya @ 12.373126
8 Ashkenazim @ 13.519056
9 Sicilian @ 14.205534
10 Greek_Cretan @ 17.168140
11 Italian-South @ 17.168957
12 Jew_Romania @ 18.039038
13 Cypriot @ 18.323494
14 Italian-Center @ 18.755230
15 Greek_East @ 19.533710
16 Jew_Syria @ 20.221943
17 Greek_Center @ 20.958565
18 Greek_South @ 22.544491
19 Ashkenazim_V @ 22.830824
20 Greek_North @ 23.659155

Using 2 populations approximation:
1 50% Samaritian +50% Sardinian @ 3.602658





Eurogenes K13:
# Population Percent
1 West_Med 37.87
2 East_Med 20.96
3 North_Atlantic 17.58
4 West_Asian 15.93
5 Baltic 4.33
6 Red_Sea 3.33


Finished reading population data. 204 populations found.
13 components mode.

--------------------------------

Least-squares method.

Using 1 population approximation:
1 West_Sicilian @ 18.553068
2 South_Italian @ 18.826334
3 Central_Greek @ 19.444452
4 Tuscan @ 19.687658
5 Italian_Abruzzo @ 19.832958
6 East_Sicilian @ 19.871048
7 Sardinian @ 21.929499
8 Greek_Thessaly @ 21.962065
9 North_Italian @ 22.072245
10 Algerian_Jewish @ 23.841873
11 Sephardic_Jewish @ 25.142250
12 Italian_Jewish @ 25.177578
13 Ashkenazi @ 25.417999
14 Spanish_Andalucia @ 27.428902
15 Spanish_Extremadura @ 28.178677
16 Bulgarian @ 28.710968
17 Tunisian_Jewish @ 29.027311
18 Libyan_Jewish @ 29.054413
19 Portuguese @ 29.754360
20 Spanish_Murcia @ 30.098209

Using 2 populations approximation:
1 50% Sardinian +50% Turkish @ 8.542939

Eurogenes K15:

# Population Percent
1 West_Med 33.93
2 East_Med 24.00
3 Atlantic 16.97
4 West_Asian 10.98
5 North_Sea 9.74
6 Red_Sea 2.42
7 Eastern_Euro 1.96


Finished reading population data. 207 populations found.
15 components mode.

--------------------------------

Least-squares method.

Using 1 population approximation:
1 Tuscan @ 16.992525
2 West_Sicilian @ 17.503353
3 South_Italian @ 18.652853
4 East_Sicilian @ 19.279425
5 Italian_Abruzzo @ 19.441479
6 Central_Greek @ 19.632788
7 North_Italian @ 19.667992
8 Sardinian @ 19.891920
9 Italian_Jewish @ 20.591984
10 Algerian_Jewish @ 20.832027
11 Greek @ 21.195177
12 Greek_Thessaly @ 21.658092
13 Ashkenazi @ 22.544605
14 Sephardic_Jewish @ 22.550343
15 Spanish_Andalucia @ 24.591578
16 Spanish_Extremadura @ 25.694555
17 Portuguese @ 26.211018
18 Spanish_Murcia @ 26.346327
19 Tunisian_Jewish @ 26.389383
20 Libyan_Jewish @ 26.890516

Using 2 populations approximation:
1 50% Sardinian +50% South_Italian @ 6.385821

EuTest:
# Population Percent
1 WEST_MED 40.25
2 EAST_MED 34.82
3 ATLANTIC 11.69
4 NORTH-CENTRAL_EURO 4.87
5 SOUTH_BALTIC 3.52
6 MIDDLE_EASTERN 2.57
7 WEST_AFRICAN 1.61


Finished reading population data. 78 populations found.
13 components mode.

--------------------------------

Least-squares method.

Using 1 population approximation:
1 Sardinian @ 26.335989
2 Tuscan @ 27.512701
3 South_Italian_&_Sicilian @ 27.724108
4 GR @ 28.300278
5 North_Italian @ 30.432611
6 AJ @ 32.538628
7 PT @ 35.293827
8 ES @ 35.843052
9 RO @ 38.581623
10 Moroccan @ 39.083221
11 Serbian @ 41.809406
12 FR @ 42.199680
13 Algerian @ 42.427990
14 Samaritan @ 42.546177
15 Druze @ 43.814205
16 TR @ 44.568321
17 AT @ 46.260983
18 HU @ 46.459305
19 French_Basque @ 46.856464
20 Armenian @ 46.924850

Using 2 populations approximation:
1 50% Druze +50% Sardinian @ 13.983733

Dodecad World9:

# Population Percent
1 Southern 45.51
2 Atlantic_Baltic 39.70
3 Caucasus_Gedrosia 13.24


Finished reading population data. 250 populations found.
9 components mode.

--------------------------------

Least-squares method.

Using 1 population approximation:
1 Sicilian_Dodecad @ 13.767551
2 Ashkenazi_Dodecad @ 14.469466
3 C_Italian_Dodecad @ 14.783624
4 Ashkenazy_Jews @ 14.981351
5 S_Italian_Sicilian_Dodecad @ 15.045189
6 S_Italian_Dodecad @ 15.521159
7 Morocco_Jews_Behar @ 16.212872
8 O_Italian_Dodecad @ 17.020256
9 Greek_Dodecad @ 17.198324
10 Sephardic_Jews_Behar @ 17.512959
11 Tuscan_HGDP @ 17.609749
12 TSI30_Metspalu @ 18.203157
13 Sardinian @ 20.207897
14 Canarias_1000 Genomes @ 22.690958
15 North_Italian_HGDP @ 23.989145
16 Murcia_1000 Genomes @ 24.609751
17 N_Italian_Dodecad @ 25.327621
18 Baleares_1000 Genomes @ 27.323318
19 Andalucia_1000 Genomes @ 27.406607
20 Bulgarian_Dodecad @ 27.564579

Using 2 populations approximation:
1 50% Sardinian +50% Morocco_Jews_Behar @ 2.613563


Using 3 populations approximation:
1 50% Sardinian +25% Tuscan_HGDP +25% Yemen_Jews_Behar @ 1.113207


Dodecad V3:
# Population Percent
1 Mediterranean 48.11
2 West_Asian 25.21
3 West_European 12.47
4 Northwest_African 6.34
5 Southwest_Asian 5.76


Finished reading population data. 227 populations found.
12 components mode.

--------------------------------

Least-squares method.

Using 1 population approximation:
1 S_Italian_Dodecad @ 6.429591
2 Sicilian_Dodecad @ 7.076639
3 S_Italian_Sicilian_Dodecad @ 8.338957
4 Sephardic_Jews_Behar @ 11.128772
5 C_Italian_Dodecad @ 11.517384
6 Greek_Dodecad @ 13.702282
7 Morocco_Jews_Behar @ 16.346556
8 Tuscan_HGDP @ 17.029627
9 O_Italian_Dodecad @ 17.076273
10 Ashkenazi_Dodecad @ 17.462656
11 Ashkenazy_Jews_Behar @ 18.817856
12 Cypriots_Behar @ 19.945402
13 Tuscan_Xing @ 22.584023
14 TSI_HapMap @ 22.936161
15 Tuscan_Henn @ 23.233191
16 North_Italian_HGDP @ 25.893709
17 N_Italian_Dodecad @ 26.113695
18 Turkish_Dodecad @ 27.622204
19 Sardinian_HGDP @ 28.457106
20 Druze_HGDP @ 29.084307

Using 2 populations approximation:
1 50% S_Italian_Dodecad +50% S_Italian_Dodecad @ 6.429591


Dodecad K7b:
# Population Percent
1 Southern 46.58
2 Atlantic_Baltic 33.93
3 West_Asian 18.13


Finished reading population data. 223 populations found.
7 components mode.

--------------------------------

Least-squares method.

Using 1 population approximation:
1 Morocco_Jews_Behar @ 10.226101
2 Ashkenazi_Dodecad @ 10.331970
3 Ashkenazy_Jews_Behar @ 10.906626
4 Sicilian_Dodecad @ 11.025823
5 S_Italian_Sicilian_Dodecad @ 11.577702
6 Sephardic_Jews_Behar @ 12.401764
7 C_Italian_Dodecad @ 15.393037
8 Greek_Dodecad @ 16.089119
9 TSI30_Metspalu @ 20.511086
10 Tuscan_HGDP @ 20.860069
11 O_Italian_Dodecad @ 23.185001
12 Cypriots_Behar @ 23.607611
13 Algerian_Dodecad @ 24.849754
14 Sardinian_HGDP @ 26.124981
15 Moroccan_Dodecad @ 26.234802
16 North_Italian_HGDP @ 27.556782
17 Canarias_1000Genomes @ 28.298319
18 N_Italian_Dodecad @ 28.830986
19 Lebanese_Behar @ 30.180227
20 Bulgarian_Dodecad @ 31.055723

Using 2 populations approximation:
1 50% Cypriots_Behar +50% Sardinian_HGDP @ 2.523127

Dodecad K12b:
# Population Percent
1 Atlantic_Med 42.97
2 Caucasus 35.18
3 Southwest_Asian 12.50
4 North_European 3.66
5 Northwest_African 3.37


Finished reading population data. 223 populations found.
12 components mode.

--------------------------------

Least-squares method.

Using 1 population approximation:
1 Sicilian_Dodecad @ 15.935287
2 S_Italian_Sicilian_Dodecad @ 16.252333
3 C_Italian_Dodecad @ 17.024044
4 Morocco_Jews_Behar @ 18.121609
5 Tuscan_HGDP @ 18.208611
6 Sephardic_Jews_Behar @ 18.281443
7 TSI30_Metspalu @ 18.942375
8 Ashkenazi_Dodecad @ 19.895327
9 Ashkenazy_Jews_Behar @ 20.798410
10 O_Italian_Dodecad @ 22.817738
11 Greek_Dodecad @ 22.880953
12 North_Italian_HGDP @ 23.574133
13 N_Italian_Dodecad @ 25.242472
14 Cypriots_Behar @ 27.593895
15 Baleares_1000Genomes @ 30.424555
16 Andalucia_1000Genomes @ 30.475300
17 Canarias_1000Genomes @ 30.959623
18 Murcia_1000Genomes @ 31.004824
19 Sardinian_HGDP @ 32.022221
20 Galicia_1000Genomes @ 32.727673

Using 2 populations approximation:
1 50% Cypriots_Behar +50% Sardinian_HGDP @ 4.011209



HarappaWorld:
# Population Percent
1 Mediterranean 39.12
2 Caucasian 36.00
3 SW-Asian 12.12
4 NE-Euro 7.54
5 Baloch 3.10
6 Pygmy 2.12


Finished reading population data. 377 populations found.
16 components mode.

--------------------------------

Least-squares method.

Using 1 population approximation:
1 morocco-jew_behar @ 15.113560
2 sephardic-jew_behar @ 15.609380
3 ashkenazy-jew_behar @ 17.855648
4 ashkenazi_harappa @ 19.632252
5 tuscan_1000genomes @ 19.884975
6 tuscan_hgdp @ 20.145630
7 tuscan_hapmap @ 20.264194
8 cypriot_behar @ 24.902407
9 italian_hgdp @ 26.690432
10 sardinian_hgdp @ 27.687870
11 turk-aydin_hodoglugil @ 29.055609
12 lebanese_behar @ 30.591202
13 turk-kayseri_hodoglugil @ 32.443691
14 lebanese-muslim_haber @ 32.940960
15 turk-istanbul_hodoglugil @ 33.070450
16 jordanian_behar @ 33.554523
17 lebanese-christian_haber @ 33.989632
18 syrian_behar @ 34.128044
19 turk_behar @ 34.129539
20 bulgarian_yunusbayev @ 34.185703

Using 2 populations approximation:
1 50% cypriot_behar +50% sardinian_hgdp @ 2.995902



PuntDNA-L K13:
# Population Percent
1 SW_Europe 50.84
2 NE_Europe 17.06
3 West_Asia 16.41
4 SW_Asia 11.99
5 Oceania 1.96
6 Americas 1.74

Using 1 population approximation:
1 Italian_Abruzzo @ 13.547079
2 Italian_Tuscan @ 14.892729
3 Italian_Sicilian @ 15.305366
4 Ashkenazy_Jew @ 16.228153
5 Italian_Bergamo @ 16.239395
6 Greek_Thessaly @ 17.058611
7 Albanian @ 17.066486
8 Greek_Central @ 17.712273
9 Kosovar @ 18.170364
10 Sephardic_Jew @ 18.244869
11 Spaniard @ 20.184280
12 Macedonian @ 22.283754
13 Bulgarian @ 22.802938
14 Puerto_Rican @ 24.422884
15 French_Basque @ 25.291721
16 Montenegrin @ 25.906849
17 Romanian @ 26.215118
18 Cypriot @ 26.631245
19 Serbian @ 27.587095
20 French @ 29.218105

Using 2 populations approximation:
1 50% Cypriot +50% French_Basque @ 7.902352


Ancient Eurasia K6:
# Population Percent
1 Natufian 50.54
2 West_European_Hunter_Gartherer 23.58
3 Ancestral_South_Eurasian 11.49
4 Ancestral_North_Eurasian 8.37
5 Sub_Saharan 6.02


Using 1 population approximation:
1 Jew_Libyan @ 13.853441
2 Jew_Tunisian @ 14.513676
3 Jew_Ashkenazi @ 15.022666
4 Jew_Moroccan @ 15.138251
5 Sicilian @ 16.429512
6 Italian_South @ 17.431561
7 Greek @ 18.535622
8 Albanian @ 19.021828
9 Libyan @ 19.204952
10 Cypriot @ 19.380522
11 Jordanian @ 19.384632
12 BedouinA @ 19.415173
13 Palestinian @ 19.510408
14 Egyptian @ 19.527651
15 Lebanese @ 19.564959
16 Saudi @ 20.471405
17 Syrian @ 20.594830
18 Armenia_ChL @ 20.724758
19 Levant_BA @ 20.959898
20 Druze @ 21.221704

Using 2 populations approximation:
1 50% Europe_EN +50% Yemeni @ 11.119516


While certain Aegean islands have absorbed such mixture and are autosomally also very close to southern Italy and western Jews, this is not the case for mainland Greece.

It is also worth noting there are some parts of Sicily and Calabria that have remained more isolated than the rest -- Messina province in the northeast, the eastern part of Palermo province, and the southern tip of Calabria -- that are genetically south/eastward of the average Ashkenazi Jew, and are often closer to Sephardim and Moroccan Jews and have a high Red Sea/SW Asian type of genetic component that is likely very ancient and predates later MENA input. While other parts of southern Italy were whacked with admixture left and right -- Berbers in southern and western Sicily, Doric Greeks in Apulia and southeast Sicily, Normans in far west Sicily, repopulation from northern Italy and so on -- this area was more or less a backwater with little colonization or foreign settlement and yet they still have a high Near Eastern component. It is also possible their extreme position on PCA plots and on calculators is due to that there was almost no Norman input there.

Greek ancestry in Italy is highest in Apulia and Lucania, as well as likely some parts of Sicily on the eastern coast such as Syracuse, but these are also not the regions that are closer to Ashkenazim/Sephardim... in fact, they deviate from the mainstream Sicily/Calabria genetic cluster on the basis of having higher Northeast European ancestry and lower Near Eastern (see my thread on the Sarno et al 2017 study, I go into great detail about all of this). Apulians and Lucanians should be descended from pre-Slavic era Greeks. This implies the Greeks settling in southern Italy already had a Northeast European component due to Indo-European settlement from the Steppe region that was higher than anything in southern Italy, but lower than in modern Greece. Yet most of the Northern admixture in Ashkenazim seems to be of WESTERN European origin, closer to what you would expect from an Italic source population.

Mycenaeans come up CLOSEST to southern Italians, Ashkenazim and Sephardim when you put them on GEDmatch or on a PCA plot, but they are still quite far from them in their components if you run them through calculators. There is no population alive today who is identical to Mycenaeans or Minoans, because every population that would have been related to them has absorbed either Levantine, Slavic, or Italic admixture. The genetic structure of Greece in general does not match either population. Both groups have significantly more Anatolian Neolithic ancestry (best proxied by Sardinians today) than any modern Greek population, and modern Greeks have significantly more Northeast European ancestry due to influence from the Slavs. Mycenaeans have too much Sardinian-like ancestry to have been the main contributor of European DNA to Ashkenazim.




Christian Lebanese should have some Aegean admixture, because the region was heavily Hellenized. I would expect Greek Orthodox Christians to be more Hellenized than Maronites.

They do not -- I have seen GEDmatch results for Greek Orthodox Lebanese and they are close to Assyrians, Druze, and so on, with no European-shifting components. Here is an example from PuntDNA-L K13:


Default Antiochian Greek Orthodox Gedmatch result!
Population Percent
1 West_Asia 33.99
2 SW_Europe 30.22
3 SW_Asia 26.99
4 NE_Europe 4.61
5 NE_Asia 1.91
6 South_Africa 0.63
7 SE_Asia 0.62
8 East_Africa 0.52
9 Oceania 0.5

Single Population Sharing:

# Population (source) Distance
1 Lebanese_Christian 3.03
2 Lebanese_Muslim 4.45
3 Lebanese_Druze 4.68
4 Syrian 4.94
5 Jordanian 5.83
6 Samaritan_Jew 7.25
7 Palestinian 7.91
8 Cypriot 8.46
9 Assyrian 10.76
10 Armenian 12.31
11 Turkish_Kayseri 12.92
12 Turkish 13.9
13 Sephardic_Jew 15.15
14 Turkish_Trabzon 16.17
15 Egyptian_Copts 16.47
16 Azerbaijan_Azeri 16.74
17 Yemeni 17.27
18 Egyptian 18.08
19 Turkish_Aydin 18.23
20 Kurdish 18.65

Mixed Mode Population Sharing:

# Primary Population (source) Secondary Population (source) Distance
1 66.5% Samaritan_Jew + 33.5% Turkish @ 2.2
2 97.6% Lebanese_Christian + 2.4% She @ 2.23
3 97.6% Lebanese_Christian + 2.4% Han_South_China @ 2.23
4 97.5% Lebanese_Christian + 2.5% Miaozu @ 2.25
5 97.7% Lebanese_Christian + 2.3% Han_North_China @ 2.26
6 97.7% Lebanese_Christian + 2.3% Korean @ 2.27
7 97.8% Lebanese_Christian + 2.2% Japanese @ 2.29
8 97.6% Lebanese_Christian + 2.4% Naxi @ 2.3
9 97.6% Lebanese_Christian + 2.4% Lahu @ 2.33
10 97.7% Lebanese_Christian + 2.3% Dai @ 2.35
11 97.6% Lebanese_Christian + 2.4% Vietnamese @ 2.38
12 97.7% Lebanese_Christian + 2.3% Xibo @ 2.38
13 97.5% Lebanese_Christian + 2.5% Burmese @ 2.43
14 97.8% Lebanese_Christian + 2.2% Daur @ 2.47
15 97.6% Lebanese_Christian + 2.4% Thai @ 2.55
16 97.9% Lebanese_Christian + 2.1% Oroqen @ 2.56
17 98.1% Lebanese_Christian + 1.9% Luzon @ 2.59
18 98% Lebanese_Christian + 2% Malayan @ 2.61
19 98.1% Lebanese_Christian + 1.9% Visayan @ 2.61
20 98.1% Lebanese_Christian + 1.9% Filipino @


Despite what people may think for cultural reasons, Christian Lebanese actually plot further from Europe than do Muslim Lebanese, because the Muslims have a small amount of Norman ancestry from the Crusades. If you put any Sicilian or Maltese through GEDMatch calculators that have both Lebanese Christians and Muslims, they will score closer to the Muslims, and the same is true for Aegean islanders.

Here is a Lebanese Muslim result on the same calculator... you can see they can be modeled with some French/Spanish type admixture, consistent with having ancestry from Normans, Italo-Normans or any other West Europeans. They do, however, have higher African input than the Christian.


# Population Percent
1 SW_Europe 30.63
2 West_Asia 30.51
3 SW_Asia 22.76
4 NE_Europe 7.29
5 South_Asia 2.46
6 East_Africa 2.22
7 West_Africa 1.18
8 Siberia 1.02
9 Americas 0.94
10 South_Africa 0.63
11 Oceania 0.37

Single Population Sharing:

# Population (source) Distance
1 Syrian 5
2 Lebanese_Christian 6.41
3 Lebanese_Druze 6.44
4 Lebanese_Muslim 6.91
5 Jordanian 7.26
6 Cypriot 7.76
7 Palestinian 9.97
8 Turkish 10.31
9 Turkish_Kayseri 10.45
10 Sephardic_Jew 11.2
11 Samaritan_Jew 12.72
12 Assyrian 13.23
13 Italian_Sicilian 14.08
14 Armenian 14.12
15 Ashkenazy_Jew 14.36
16 Turkish_Aydin 14.67
17 Azerbaijan_Azeri 15.47
18 Turkish_Trabzon 16.94
19 Egyptian 17.39
20 Egyptian_Copts 18.16

Mixed Mode Population Sharing:

# Primary Population (source) Secondary Population (source) Distance
1 91.1% Syrian + 8.9% French_Basque @ 2.14
2 88.4% Syrian + 11.6% Puerto_Rican @ 2.27
3 89.5% Syrian + 10.5% Spaniard @ 2.28
4 71.3% Syrian + 28.7% Sephardic_Jew @ 2.36
5 75.9% Syrian + 24.1% Italian_Sicilian @ 2.37
6 85.1% Lebanese_Druze + 14.9% Puerto_Rican @ 2.44
7 88.2% Syrian + 11.8% Italian_Bergamo @ 2.47
8 76.6% Syrian + 23.4% Ashkenazy_Jew @ 2.52
9 81.7% Syrian + 18.3% Italian_Abruzzo @ 2.54
10 86.1% Syrian + 13.9% Italian_Tuscan @ 2.6
11 85.4% Syrian + 14.6% Albanian @ 2.78
12 91.1% Syrian + 8.9% French @ 2.82
13 85.6% Lebanese_Christian + 14.4% Puerto_Rican @ 2.84
14 85.3% Syrian + 14.7% Greek_Thessaly @ 2.86
15 91.2% Syrian + 8.8% Colombian @ 2.9
16 84.4% Lebanese_Muslim + 15.6% Puerto_Rican @ 2.91
17 86.3% Syrian + 13.7% Kosovar @ 2.94
18 81.8% Syrian + 18.2% Greek_Central @ 2.97
19 91.8% Syrian + 8.2% Belgian @ 3
20 88.3% Syrian + 11.7% Macedonian @ 3.03




I wonder, how do people explain admixture of Levantines among modern Greek Islanders, while tend to reject the notion of Greek admixture with Levantines, when in fact Greek colonization in the Eastern Mediterranean was very prominent from at least the 10th century BC, onward.

Greek islanders have Levantine ancestry, not the reverse (this is one reason they are part of the southern Italian genetic cluster and deviate from mainland Greece, in addition to the Slavic ancestry in the mainland). Greek islanders and mainlanders are not only different because the latter have Slavic ancestry, but also because the former have some Levantine.



There's no compelling evidence in favour of Levantine admixture among Aegean Greeks, if that's what you're talking about.

Their autosomal plotting would suggest they have it, even though there is no historical record of a massive Levantine settlement in the islands.



Oh but I do ! I do believe Copts should appear to be admixed with Greeks. Actually, I would believe many Copts would probably have higher admixture with ancient Greeks than Christian Lebanese.

Neither group has Greek admixture. Copts especially do not, and I can provide autosomal results from GEDmatch to prove it. What you are not understanding about ancient Greek colonization is that they primarily assimilated people culturally, not through demographic replacement.

This is the GEDmatch for a Copt on PuntDNA-L K13:


# Population Percent
1 SW_Asia 38.2
2 SW_Europe 26.19
3 West_Asia 24.49
4 East_Africa 9.62
5 NE_Asia 1.04
6 SE_Asia 0.19
7 Oceania 0.14
8 South_Africa 0.14

Single Population Sharing:

# Population (source) Distance
1 Egyptian_Copts 2.38
2 Egyptian 10.65
3 Yemeni 11.39
4 Palestinian 11.94
5 Jordanian 13.59
6 Saudi 14.81
7 Samaritan_Jew 15.24
8 Syrian 18.46
9 Lebanese_Christian 18.78
10 Lebanese_Muslim 19.26
11 Lebanese_Druze 19.74
12 Bedouin 21.15
13 Tunisian 22.29
14 Moroccan 23.24
15 Cypriot 23.39
16 Algerian 24.56
17 Sephardic_Jew 25.41
18 Assyrian 25.57
19 Saharawi 26.93
20 Turkish_Kayseri 27.57

Mixed Mode Population Sharing:

# Primary Population (source) Secondary Population (source) Distance
1 96% Egyptian_Copts + 4% Amhara @ 1.67
2 95.8% Egyptian_Copts + 4.2% Tigray @ 1.7
3 96.9% Egyptian_Copts + 3.1% Somali @ 1.72
4 96.8% Egyptian_Copts + 3.2% Oromo @ 1.74
5 97.1% Egyptian_Copts + 2.9% Wolayta @ 1.8
6 97.6% Egyptian_Copts + 2.4% Aricultivator @ 1.84
7 97.5% Egyptian_Copts + 2.5% Datog @ 1.9
8 98.4% Egyptian_Copts + 1.6% Gumuz @ 1.97
9 97.9% Egyptian_Copts + 2.1% Maasai @ 2.02
10 98.7% Egyptian_Copts + 1.3% Han_North_China @ 2.1
11 98.7% Egyptian_Copts + 1.3% She @ 2.1
12 92.9% Egyptian_Copts + 7.1% Saudi @ 2.1
13 98.7% Egyptian_Copts + 1.3% Dinka @ 2.11
14 98.7% Egyptian_Copts + 1.3% Korean @ 2.11
15 98.7% Egyptian_Copts + 1.3% Han_South_China @ 2.11
16 98.8% Egyptian_Copts + 1.2% Japanese @ 2.11
17 98.7% Egyptian_Copts + 1.3% Naxi @ 2.11
18 98.7% Egyptian_Copts + 1.3% Miaozu @ 2.11
19 98.7% Egyptian_Copts + 1.3% Anuak @ 2.12
20 98.6% Egyptian_Copts + 1.4% Burmese @ 2.13




The reason why modern Western Jews cluster so close with South Italians, Sicilians and Islander Greeks, is because all of these locations have substantial ancient Greek ancestry. Furthermore, as a result of Sicilians being a mixture of ancient Greeks and Levantines (from Phoenician, Punic and then Arabic rule), and a small amount of recent Northern European admixture, they score closest to Western Jews.

Modern Greek Islanders are closer to Sicilians than to modern Mainland Greeks. Modern mainland Greeks are admixed with Early Medieval Slavs, which entered the Balkans. Modern Greek Islanders have kept isolated from that to a degree, and also got more recent Levantine admixture (which make them appear close to Sicilians and Western Jews, again).

These are contradictory statements, but you are somewhat correct. Sicilians, Aegean islanders, Jews, Calabrese, etc. plot together not because of ancient Greek ancestry but because they share common Levantine ancestry, which may or may not predate Phoenicians. This is why they form their own cluster, away from other Italians and Greeks. I know I keep repeating myself but Greek islanders are not isolated as you might think either. Several islands were repopulated with Greeks from Asia Minor and Cyprus who would have carried with them higher Near Eastern components. Some islands, like the Cyclades and North Aegean, shift toward mainland Greece also because of greater settlement from there.

Erikl86
06-08-2018, 07:56 AM
Sikeliot, thanks for the in-depth response :).




Western Jews descend from a population of Jews who lived primarily in central and southern Italy and moved northward, not in Greece. They were Greek-speaking because southern Italy as a whole was Greek-speaking, but Greece itself in that era did not have the largest Jewish population in Europe;


We agree - I agree that most probably, the mutual forefathers of Western Jews arrived to France via Italy. However, I've never claimed "Greece", as in mainland Greece, had the largest Jewish population. I was talking about the Greek "half" of the Roman Empire - mainly Eastern Mediterranean and Egypt. By far, until the 2-3rd centuries, it had the largest Jewish population. Most proselytizes happened in that part of the Roman Empire.



Italy did. Italkim are probably a good proxy group in Italy for one of the older Jewish groups, who would have migrated north, acquired significant Italic admixture, and then spread into Europe.

Yes, but Italkim's rites are extremely similar, and probably derive from, Romanoite Jews - the oldest Jewish community in Greek. You also had small number of Romanoite Jews still living in S. Italy to modern times. It's quite obvious, that Italkim derive from this community.


It is very difficult to distinguish modern Ashkenazim/Sephardim from southern Italians on an autosomal level today because these parts of Italy have ancient Levantine ancestry that is likely very close to that of the Canaanite populations who became Phoenicians and Jews.

Again, this is part of what I'm saying - no argument here. Phoenicians have been settling in S. Italy, Sicily and Malta from at least the 11th century BC, if not before. Sicily was also part of Carthage for few centuries.


However, due to that this admixture does not directly correspond with the parts of southern Italy that received Phoenician, Carthaginian, and Arab/Berber admixture, we should view at least part of this as very ancient admixture that would have been present since very ancient times.

Not that ancient - I've linked here to an article that studied just that - the time and place when such admixtures happened among Sicilians, S. Italians, and Greeks.

Most Near Eastern admixture correspond to the Phoenician arrival to Sicily. If it was much older, then you'd see similar findings in nearby Sardinia. Which you don't.


There is a Bronze Age Sicilian GEDmatch result which already has such admixture in significant amounts:


If you refer to the samples from a "Beaker" burial site at Partanna, western Sicily, and dated to 2500–1900 BCE, then your results seem like an outlier results. Look at what I got, at least for oracle:


K13 Oracle ref data revised 21 Nov 2013

Admix Results (sorted):

# Population Percent
1 West_Med 37.87
2 East_Med 20.96
3 North_Atlantic 17.58
4 West_Asian 15.93
5 Baltic 4.33
6 Red_Sea 3.33


Finished reading population data. 204 populations found.
13 components mode.

--------------------------------

Least-squares method.

Using 1 population approximation:
1 West_Sicilian @ 18.553068
2 South_Italian @ 18.826334
3 Central_Greek @ 19.444452
4 Tuscan @ 19.687658
5 Italian_Abruzzo @ 19.832958
6 East_Sicilian @ 19.871048
7 Sardinian @ 21.929499
8 Greek_Thessaly @ 21.962065
9 North_Italian @ 22.072245
10 Algerian_Jewish @ 23.841873
11 Sephardic_Jewish @ 25.142250
12 Italian_Jewish @ 25.177578
13 Ashkenazi @ 25.417999
14 Spanish_Andalucia @ 27.428902
15 Spanish_Extremadura @ 28.178677
16 Bulgarian @ 28.710968
17 Tunisian_Jewish @ 29.027311
18 Libyan_Jewish @ 29.054413
19 Portuguese @ 29.754360
20 Spanish_Murcia @ 30.098209

Using 2 populations approximation:
1 50% Sardinian +50% Turkish @ 8.542939


Using 3 populations approximation:
1 50% Sardinian +25% Sardinian +25% Tabassaran @ 6.007061


Using 4 populations approximation:
++++++++++++++++++++++++++++++++++++++++++++++++++ +++++++++++++++++++++++++++++++++++++
1 Abhkasian + North_Italian + Sardinian + Sardinian @ 4.008021
2 Georgian + Sardinian + Sardinian + Tuscan @ 4.097898
3 Abhkasian + Sardinian + Sardinian + Tuscan @ 4.144988
4 Georgian + North_Italian + Sardinian + Sardinian @ 4.230395
5 Ossetian + Sardinian + Sardinian + South_Italian @ 4.520500
6 Georgian + Sardinian + Sardinian + West_Sicilian @ 4.540919
7 Ossetian + Sardinian + Sardinian + West_Sicilian @ 4.600028
8 Central_Greek + Ossetian + Sardinian + Sardinian @ 4.745520
9 East_Sicilian + Ossetian + Sardinian + Sardinian @ 4.773656
10 Balkar + Sardinian + Sardinian + South_Italian @ 4.800226
11 Abhkasian + Sardinian + Sardinian + West_Sicilian @ 4.886259
12 Georgian + Italian_Abruzzo + Sardinian + Sardinian @ 4.934696
13 Kabardin + Sardinian + Sardinian + South_Italian @ 4.968035
14 Abhkasian + Sardinian + Sardinian + Sardinian @ 4.992015
15 Abhkasian + Sardinian + Sardinian + Spanish_Andalucia @ 5.023455
16 Italian_Abruzzo + Ossetian + Sardinian + Sardinian @ 5.040259
17 Balkar + Sardinian + Sardinian + West_Sicilian @ 5.045648
18 Adygei + Sardinian + Sardinian + South_Italian @ 5.049405
19 Ossetian + Sardinian + Sardinian + Tuscan @ 5.056332
20 Abhkasian + Sardinian + Sardinian + Spanish_Extremadura @ 5.059998

Also, compare modern Sicilians to Bronze Age Sicilians:


[1] distance%=5.0349 / distance=0.050349

Beaker_Sicily:I4930

Peloponnese_N 60.55
Peloponnese_N_outlier:I3920 17.6
Armenia_EBA 15.3
WHG 6.25
Moroccan 0.3

[1] distance%=2.3039 / distance=0.023039

Sicilian_East

Peloponnese_N_outlier:I3920 47.9
Yamnaya_Samara 17.8
Armenia_EBA 9.8
Levant_N 8.25
Barcin_N 6.55
Moroccan 6.4
WHG 3.3

[1] distance%=2.1998 / distance=0.021998

Sicilian_West

Peloponnese_N_outlier:I3920 52.9
Yamnaya_Samara 17.05
Moroccan 8.55
Levant_N 8.25
Armenia_EBA 7
WHG 6.25

And you can also check the PCA map where these samples cluster:

23784

Compare that, to BA Greeks (Mycenaeans and Minoans):


Mycenaean sample (I9041)

K13 Oracle ref data revised 21 Nov 2013

Admix Results (sorted):

# Population Percent
1 East_Med 34.14
2 West_Med 30.04
3 North_Atlantic 20.94
4 West_Asian 9.36
5 Red_Sea 3.74
6 Siberian 1.21


Finished reading population data. 204 populations found.
13 components mode.

--------------------------------

Least-squares method.

Using 1 population approximation:
1 South_Italian @ 12.130267
2 West_Sicilian @ 12.197648
3 Italian_Jewish @ 12.297517
4 Algerian_Jewish @ 12.913167
5 Sephardic_Jewish @ 13.605566
6 East_Sicilian @ 15.037803
7 Italian_Abruzzo @ 15.824724
8 Central_Greek @ 16.181517
9 Tuscan @ 16.462431
10 Ashkenazi @ 16.676908
11 Tunisian_Jewish @ 17.475962
12 Libyan_Jewish @ 17.959299
13 Greek_Thessaly @ 21.088852
14 North_Italian @ 22.279140
15 Cyprian @ 23.118532
16 Tunisian @ 24.614328
17 Algerian @ 25.246191
18 Mozabite_Berber @ 26.188852
19 Moroccan @ 27.004148
20 Sardinian @ 27.060648

Using 2 populations approximation:
1 50% Lebanese_Druze +50% Sardinian @ 7.982687


Using 3 populations approximation:
1 50% Italian_Jewish +25% Sardinian +25% Sephardic_Jewish @ 5.438547


Using 4 populations approximation:
++++++++++++++++++++++++++++++++++++++++++++++++++ +++++++++++++++++++++++++++++
1 Italian_Jewish + Italian_Jewish + Sardinian + Sephardic_Jewish @ 5.438547
2 Italian_Jewish + Italian_Jewish + Italian_Jewish + Sardinian @ 5.443801
3 Italian_Jewish + Italian_Jewish + Sardinian + South_Italian @ 5.461736
4 Italian_Jewish + Lebanese_Druze + Sardinian + Spanish_Andalucia @ 5.573959
5 Italian_Jewish + Italian_Jewish + Sardinian + West_Sicilian @ 5.605775
6 Italian_Jewish + Sardinian + Sephardic_Jewish + South_Italian @ 5.609292
7 Italian_Jewish + Sardinian + Sephardic_Jewish + Sephardic_Jewish @ 5.613409
8 Italian_Jewish + Sardinian + Sephardic_Jewish + West_Sicilian @ 5.684255
9 French_Basque + Lebanese_Christian + Lebanese_Druze + Sardinian @ 5.708206
10 Italian_Abruzzo + Italian_Jewish + Italian_Jewish + Sardinian @ 5.750822
11 Italian_Jewish + Lebanese_Christian + Sardinian + Spanish_Andalucia @ 5.767160
12 French_Basque + Lebanese_Druze + Lebanese_Druze + Sardinian @ 5.802987
13 Lebanese_Druze + Sardinian + Spanish_Andalucia + Tunisian_Jewish @ 5.806496
14 Lebanese_Druze + Sardinian + Sephardic_Jewish + Spanish_Andalucia @ 5.890898
15 Sardinian + Sephardic_Jewish + Sephardic_Jewish + South_Italian @ 5.909827
16 Sardinian + Sephardic_Jewish + Sephardic_Jewish + West_Sicilian @ 5.914345
17 Lebanese_Druze + Sardinian + Spanish_Aragon + Tunisian_Jewish @ 5.931470
18 Sardinian + Sephardic_Jewish + Sephardic_Jewish + Sephardic_Jewish @ 5.943864
19 Italian_Abruzzo + Italian_Jewish + Sardinian + Sephardic_Jewish @ 5.951684
20 Italian_Jewish + Lebanese_Druze + Sardinian + Spanish_Aragon @ 5.977618

And as for when modern Greeks got their Levantine admixture:

23785

And, this is how Lazaridis et al. qpAdm model re-constructed:


Mycenaean
Minoan_Lasithi 0.786±0.049
Sintashta 0.214±0.049
P-value 0.96574059
chisq 6.030
Full output

Mycenaean
Corded_Ware_Germany 0.210±0.043
Minoan_Lasithi 0.790±0.043
P-value 0.961238695
chisq 6.198
Full output

Mycenaean
Minoan_Lasithi 0.791±0.043
Srubnaya 0.209±0.043
P-value 0.950419642
chisq 6.558
Full output

And also, PCA map:

23786

Bronze Age Sicilians, which are roughly from the same period as Bronze Age Greeks, seem to be much, much further away from modern Jews or Levantines in general, while the Bronze Age Greek got Western Jews much closer to him (and also, other Levantines like Druze), with Sardinian admixture. Being that Mycenaeans had little to no Levantine admixture, as shown, this is most probably due to the non-Levantine, south European component, resembling the same component that Western Jews has.



While certain Aegean islands have absorbed such mixture and are autosomally also very close to southern Italy and western Jews, this is not the case for mainland Greece.

Agreed.




It is also worth noting there are some parts of Sicily and Calabria that have remained more isolated than the rest -- Messina province in the northeast, the eastern part of Palermo province, and the southern tip of Calabria -- that are genetically south/eastward of the average Ashkenazi Jew, and are often closer to Sephardim and Moroccan Jews and have a high Red Sea/SW Asian type of genetic component that is likely very ancient and predates later MENA input.
Red Sea and SW Asian in Sicily? most unlikely to be ancient - most likely Phoenician.


While other parts of southern Italy were whacked with admixture left and right -- Berbers in southern and western Sicily, Doric Greeks in Apulia and southeast Sicily, Normans in far west Sicily, repopulation from northern Italy and so on -- this area was more or less a backwater with little colonization or foreign settlement and yet they still have a high Near Eastern component. It is also possible their extreme position on PCA plots and on calculators is due to that there was almost no Norman input there.

Impossible to tell. The isolation might have caused genetic drift, like we see in Samaritans or Ashkenazi Jews. But I would find it extremely difficult to believe that such high level of clearly, Near Eastern admixture in Sicily and S. Italy of all the places, is not due to Phoenicians, Carthaginians or Arabs.


Greek ancestry in Italy is highest in Apulia and Lucania, as well as likely some parts of Sicily on the eastern coast such as Syracuse, but these are also not the regions that are closer to Ashkenazim/Sephardim... in fact, they deviate from the mainstream Sicily/Calabria genetic cluster on the basis of having higher Northeast European ancestry and lower Near Eastern (see my thread on the Sarno et al 2017 study, I go into great detail about all of this). Apulians and Lucanians should be descended from pre-Slavic era Greeks. This implies the Greeks settling in southern Italy already had a Northeast European component due to Indo-European settlement from the Steppe region that was higher than anything in southern Italy, but lower than in modern Greece.
Not accurate - Apulia and Lucania recieved Greek immigrants all through Medieval Ages. As can be seen from the table I've posted above, Slavic admixture started out in mainland Greece from around the 5-6th century AD. Many places in S. Italy and Sicily were under Byzantine rule all the way to the 10th century, and also after the collapse of the Byzantine Empire, new waves have arrived from mainland Greece and settled in locations that had Greek-speakers, like Calabria:


Although many of the Greek inhabitants of Southern Italy were entirely Italianized during the Middle Ages (for example, Paestum was by the 4th century BC), pockets of Greek culture and language remained and survived into modernity partly because of continuous migration between southern Italy and the Greek mainland. One example is the Griko people, some of whom still maintain their Greek language and customs.[citation needed]

Greeks re-entered the region in the 16th and 17th century in reaction to the conquest of the Peloponnese by the Ottoman Empire. Especially after the end of the Siege of Coron (1534), large numbers of Greeks took refuge in the areas of Calabria, Salento and Sicily. Greeks from Coroni, the so-called Coronians, were nobles, who brought with them substantial movable property. They were granted special privileges and tax exemptions.

Other Greeks who moved to Italy came from the Mani Peninsula of the Peloponnese. The Maniots were known for their proud military traditions and for their bloody vendettas, many of which still continue today. Another group of Maniot Greeks moved to Corsica
Source: https://en.wikipedia.org/wiki/Magna_Graecia

This would, indeed, infuse the local Italian Greeks with Greeks that had already were admixed with Slavs.


Yet most of the Northern admixture in Ashkenazim seems to be of WESTERN European origin, closer to what you would expect from an Italic source population.
I'd agree it Ashkenazi Northern European admixture comes from Western European - most likely French/Rhineland Germans.
You seem to contradict yourself here a little bit.


Mycenaeans come up CLOSEST to southern Italians, Ashkenazim and Sephardim when you put them on GEDmatch or on a PCA plot, but they are still quite far from them in their components if you run them through calculators. There is no population alive today who is identical to Mycenaeans or Minoans, because every population that would have been related to them has absorbed either Levantine, Slavic, or Italic admixture. The genetic structure of Greece in general does not match either population. Both groups have significantly more Anatolian Neolithic ancestry (best proxied by Sardinians today) than any modern Greek population, and modern Greeks have significantly more Northeast European ancestry due to influence from the Slavs.
Agreed 100%. This is why I never just use ancient Aegeans - that would be weird. But if the major admixture of Western Jews were Mycenaean-like Greeks (that is, ancient Greeks which would be much closer genetically to BA Mycenaean than modern Greeks), and after that they had little admixture with other southern Europeans, and they give high admixture and good distance in modern Western Jews (as well as Druze), then again, the fact that such ancient samples give good results, pretty much shows that it's probably where Jews got their Southern European admixture from.


Mycenaeans have too much Sardinian-like ancestry to have been the main contributor of European DNA to Ashkenazim.
Less so than BA Sicilians, as I've shown.





They do not -- I have seen GEDmatch results for Greek Orthodox Lebanese and they are close to Assyrians, Druze, and so on, with no European-shifting components. Here is an example from PuntDNA-L K13:


Default Antiochian Greek Orthodox Gedmatch result!
Population Percent
1 West_Asia 33.99
2 SW_Europe 30.22
3 SW_Asia 26.99
4 NE_Europe 4.61
5 NE_Asia 1.91
6 South_Africa 0.63
7 SE_Asia 0.62
8 East_Africa 0.52
9 Oceania 0.5

Single Population Sharing:

# Population (source) Distance
1 Lebanese_Christian 3.03
2 Lebanese_Muslim 4.45
3 Lebanese_Druze 4.68
4 Syrian 4.94
5 Jordanian 5.83
6 Samaritan_Jew 7.25
7 Palestinian 7.91
8 Cypriot 8.46
9 Assyrian 10.76
10 Armenian 12.31
11 Turkish_Kayseri 12.92
12 Turkish 13.9
13 Sephardic_Jew 15.15
14 Turkish_Trabzon 16.17
15 Egyptian_Copts 16.47
16 Azerbaijan_Azeri 16.74
17 Yemeni 17.27
18 Egyptian 18.08
19 Turkish_Aydin 18.23
20 Kurdish 18.65

Mixed Mode Population Sharing:

# Primary Population (source) Secondary Population (source) Distance
1 66.5% Samaritan_Jew + 33.5% Turkish @ 2.2
2 97.6% Lebanese_Christian + 2.4% She @ 2.23
3 97.6% Lebanese_Christian + 2.4% Han_South_China @ 2.23
4 97.5% Lebanese_Christian + 2.5% Miaozu @ 2.25
5 97.7% Lebanese_Christian + 2.3% Han_North_China @ 2.26
6 97.7% Lebanese_Christian + 2.3% Korean @ 2.27
7 97.8% Lebanese_Christian + 2.2% Japanese @ 2.29
8 97.6% Lebanese_Christian + 2.4% Naxi @ 2.3
9 97.6% Lebanese_Christian + 2.4% Lahu @ 2.33
10 97.7% Lebanese_Christian + 2.3% Dai @ 2.35
11 97.6% Lebanese_Christian + 2.4% Vietnamese @ 2.38
12 97.7% Lebanese_Christian + 2.3% Xibo @ 2.38
13 97.5% Lebanese_Christian + 2.5% Burmese @ 2.43
14 97.8% Lebanese_Christian + 2.2% Daur @ 2.47
15 97.6% Lebanese_Christian + 2.4% Thai @ 2.55
16 97.9% Lebanese_Christian + 2.1% Oroqen @ 2.56
17 98.1% Lebanese_Christian + 1.9% Luzon @ 2.59
18 98% Lebanese_Christian + 2% Malayan @ 2.61
19 98.1% Lebanese_Christian + 1.9% Visayan @ 2.61
20 98.1% Lebanese_Christian + 1.9% Filipino @


Your first fits are with other Lebanese, which I'd agree are closer (and have shown here as well that they are all pretty much admixed with the same populations), your first non-Levantine closest single population are Cypriots - which can be modelled as 63% Ancient Aegean!

Later on, your best Oracle fit is Samaritan (66%) + Turkish (33%) - and modern Turks are descendants of Hellenized Byazntine Anatolians, an area which has been heavily colonized by Greeks since the 10th century BC all the way to the 15th century AD. You actually proved the model I've shown for them:


[1] "distance%=0.0292 / distance=0.000292"

Lebanese_Christian:average

Minoan_Lasithi:I0070 31.55
Iran_ChL:I1661 25.15
Levant_N:I1704 24.50
Iran_ChL:I1665 12.55
Mycenaean:I9041 4.30
Levant_N:I1699 1.65
Iran_ChL:I1662 0.30

Roughly 65% Bronze Age Levantines, 35% ancient Aegean. My bet, these are Greek Orthodox Christian Lebanese which have been sampled

Distance of 0.0292%.

The same case for Muslim Lebanese:


Single Population Sharing:

# Population (source) Distance
1 Syrian 5
2 Lebanese_Christian 6.41
3 Lebanese_Druze 6.44
4 Lebanese_Muslim 6.91
5 Jordanian 7.26
6 Cypriot 7.76
7 Palestinian 9.97
8 Turkish 10.31
9 Turkish_Kayseri 10.45
10 Sephardic_Jew 11.2
11 Samaritan_Jew 12.72
12 Assyrian 13.23
13 Italian_Sicilian 14.08
14 Armenian 14.12
15 Ashkenazy_Jew 14.36
16 Turkish_Aydin 14.67
17 Azerbaijan_Azeri 15.47
18 Turkish_Trabzon 16.94
19 Egyptian 17.39
20 Egyptian_Copts 18.16


This is how Cypriots can be modelled:


[1] "distance%=0.0998 / distance=0.000998"

Cypriot:average

Minoan_Odigitria:I9131 24.80
Mycenaean:I9041 21.40
Minoan_Lasithi:I0073 17.10
Iran_ChL:I1661 12.00
Iran_ChL:I1662 8.65
Iran_ChL:I1665 7.25
Levant_N:I1704 6.70
Iran_ChL:I1674 2.10

Mirror image of Druze . 63% Aegean, 37% Levantines.

And also, Lazaridis et al. study shows this is accurate as well:

23787

Greek Cypriots (most to the right column) get more than 20% Levantine.



Greek islanders have Levantine ancestry, not the reverse (this is one reason they are part of the southern Italian genetic cluster and deviate from mainland Greece, in addition to the Slavic ancestry in the mainland). Greek islanders and mainlanders are not only different because the latter have Slavic ancestry, but also because the former have some Levantine.

Agreed, and I've stated it before - Greek Islanders are much more Levantine mixed than mainland Greece. Could be for several reasons, but IMO it's either simple distance made the Aegeans closer to the Levant during Hellenistic times (when most admixture with Levantines occurred) or simply because the large Slavic influx in mainland Greece "pushed" down the admixture.




Their autosomal plotting would suggest they have it, even though there is no historical record of a massive Levantine settlement in the islands.



There need not to be any specific historical record - Greek-centered empires have been ruling the Levant since at least Alexander the Great, until Islam. The populations mixed.


Neither group has Greek admixture. Copts especially do not, and I can provide autosomal results from GEDmatch to prove it. What you are not understanding about ancient Greek colonization is that they primarily assimilated people culturally, not through demographic replacement.


I have yet to examine Copts, so I would have to take your word for it. However, I'd find it very hard to believe, given the extensive colonization and the Greek city of Alexandria.

Druze, Lebanese and even Palestinian Christians seem to have some Aegean admixture (20-30%).




These are contradictory statements, but you are somewhat correct. Sicilians, Aegean islanders, Jews, Calabrese, etc. plot together not because of ancient Greek ancestry but because they share common Levantine ancestry, which may or may not predate Phoenicians. This is why they form their own cluster, away from other Italians and Greeks.


I say for both - and I really can't say you managed to prove otherwise. Except for the BA Sicilian outlier, most of what you said doesn't manage to prove otherwise - plus you assumed few things which were simply wrong, historically, as I've shown here otherwise (like the fact that mainland Greece with Slavic admixture didn't immigrate to S. Italy).


I know I keep repeating myself but Greek islanders are not isolated as you might think either. Several islands were repopulated with Greeks from Asia Minor and Cyprus who would have carried with them higher Near Eastern components. Some islands, like the Cyclades and North Aegean, shift toward mainland Greece also because of greater settlement from there.

I've never said they are - they are just more isolated than mainland Greece, especially when it comes to Slavic admixture.

eolien
06-08-2018, 08:57 AM
So many remarks...

I cannot imagine that Greek islanders hsving real Levantine ancestry, what they have is the Anatolian Farmer component which unites us all (I think that was Sikeliot's point?). As far as I can see some of you guys are mixing up statistical significance with historical reality. Life is not so easy, at least not for me. Whatever mathematical method we use, as a molecular biologist I can see that the data is derived from SNPs and assigned to populations. However most SNPs are common among historically recent populations. For example both my parents show around 40% European farmer component in "Kurd"s K14 geneplaza test. One of the highests among the tested people so far. But as far as I can see, he uses the distinguishing SNPs instead. Even if the european jews were descendants from Greeks still most Jews lived in Anatolia and Thrace at that time and much less in Greece and in the islands proper. How close were the Hellenized populations to the Greeks genetically?

Still when i saw the G25 PCA of Davidski, i was really surprised that both Ashkenazi and Sefardic samples appearso close to Eastern med. and distinct from Iraqi Jews or Samaritans. I would have expected Ashkenazi more shifted to the north due to supposed admixture and Sefardi closer to Spain or to North African Jews etc. In fact they look more southern in the PCA (eastern in reality) than the Italians. This cannot be a coincidence. The Greeks are not there alone but also Anatolians and Lebanese Christians are closeby etc. The Sea people might have come from the islands, or cilicia or cyprus. In any case they mixed very soon with the local population.

Finally How do samaritans represent an ancient native population? Was there no conversion to Samartian religion, for me they are a relic of something that included many converts and settlers from Assyria.

There are multiple scenarios that could be statistically fit and logical but difficult to know which one(s) are real.


23788

Sikeliot
06-08-2018, 11:24 AM
The issue is how you are labeling ancestral components, it is causing you to draw faulty conclusions.



We agree - I agree that most probably, the mutual forefathers of Western Jews arrived to France via Italy. However, I've never claimed "Greece", as in mainland Greece, had the largest Jewish population. I was talking about the Greek "half" of the Roman Empire - mainly Eastern Mediterranean and Egypt. By far, until the 2-3rd centuries, it had the largest Jewish population. Most proselytizes happened in that part of the Roman Empire.

You said "Aegean" but again there is no proof that this is the case nor, as has been demonstrated, is there an "Aegean" genetic profile. Because clearly as we see, Greek mainlanders are one genetic cluster, Aegean islanders are in the southern Italian cluster, and modern day Turks and Cypriots are distinct from both. It can also not be proven conclusively that in ancient times, even before Slavs and other groups migrated to the area that there were not also internal differences between mainland and islands.




Yes, but Italkim's rites are extremely similar, and probably derive from, Romanoite Jews - the oldest Jewish community in Greek. You also had small number of Romanoite Jews still living in S. Italy to modern times. It's quite obvious, that Italkim derive from this community.

Or it could be the other way around. Or both groups could simply have older customs because they are earlier established communities that remained isolated and possibly in contact with one another. Again, we do not know.



Again, this is part of what I'm saying - no argument here. Phoenicians have been settling in S. Italy, Sicily and Malta from at least the 11th century BC, if not before. Sicily was also part of Carthage for few centuries.

I do think that Phoenician admixture explains some of what is going on as well as other groups that were in Sicily like Elymians (relatives of the Hittites who would have carried high Caucasus type ancestry), but I think that you still see some degree of Near Eastern admixture in places where there was no historical Phoenician nor Carthaginian settlement shows that some of it is more ancient than that.



Not that ancient - I've linked here to an article that studied just that - the time and place when such admixtures happened among Sicilians, S. Italians, and Greeks.

That study stated that some of the admixture is recent, but it didn't state that all of it is.


Most Near Eastern admixture correspond to the Phoenician arrival to Sicily. If it was much older, then you'd see similar findings in nearby Sardinia. Which you don't.

It could be an accumulation of multiple admixture events but some of it is likely to be more ancient than that.




If you refer to the samples from a "Beaker" burial site at Partanna, western Sicily, and dated to 2500–1900 BCE, then your results seem like an outlier results. Look at what I got, at least for oracle:



Also, compare modern Sicilians to Bronze Age Sicilians:



And you can also check the PCA map where these samples cluster:

23784

Compare that, to BA Greeks (Mycenaeans and Minoans):



And as for when modern Greeks got their Levantine admixture:

23785

And, this is how Lazaridis et al. qpAdm model re-constructed:



And also, PCA map:

23786

Bronze Age Sicilians, which are roughly from the same period as Bronze Age Greeks, seem to be much, much further away from modern Jews or Levantines in general, while the Bronze Age Greek got Western Jews much closer to him (and also, other Levantines like Druze), with Sardinian admixture. Being that Mycenaeans had little to no Levantine admixture, as shown, this is most probably due to the non-Levantine, south European component, resembling the same component that Western Jews has.


We also do not know for a fact that result represents all of Sicily's population in its era. The Beaker culture came to Sicily from Sardinia and the Sarno et al study you have quoted states that Trapani (which is where Partanna is located) differs from the rest of Sicily due to two factors: higher East European admixture (which could be Greek) and higher Sardinian-like admixture which is recent. This could be the Beakers. If you took a sample from the same era from a different part of Sicily you would likely find something resembling Minoans considering that the sample from Partanna was basically similar to Minoans with extra Sardinian admixture. But within that they still had SOME degree of Near Eastern ancestry, showing it is not entirely from Phoenicians and Arabs even if some of it definitely is. Modern Sicilians are more Near Eastern than that sample, but some of that ancestry did already exist there.



Not accurate - Apulia and Lucania recieved Greek immigrants all through Medieval Ages. As can be seen from the table I've posted above, Slavic admixture started out in mainland Greece from around the 5-6th century AD. Many places in S. Italy and Sicily were under Byzantine rule all the way to the 10th century, and also after the collapse of the Byzantine Empire, new waves have arrived from mainland Greece and settled in locations that had Greek-speakers, like Calabria:


Source: https://en.wikipedia.org/wiki/Magna_Graecia

This would, indeed, infuse the local Italian Greeks with Greeks that had already were admixed with Slavs.

Considering that most of Sicily and Calabria have no IBD sharing with mainland Greece and very low NE European admixture (unlike Apulia which has higher IBD sharing with Greece and Eastern Europe -- see the Sarno study again), it is unlikely there was a recent mass migration of mainland Greeks to Calabria. In fact of all southern Italians, the people of southern Calabria have the oldest DNA and they have mixed very little with non-locals. The Greek speaking population especially, but they still do not show IBD sharing with modern Greeks which might suggest that they acquired the Greek language very early on and then did not see any recent infusion of foreign DNA.



I'd agree it Ashkenazi Northern European admixture comes from Western European - most likely French/Rhineland Germans.
You seem to contradict yourself here a little bit.

The suggestion is that the non-Levantine DNA in Ashkenazim is from northern Italy. Northern Italians share some degree of ancestry with French and Germans, especially considering they have both Gallic and Germanic ancestry.



Your first fits are with other Lebanese, which I'd agree are closer (and have shown here as well that they are all pretty much admixed with the same populations), your first non-Levantine closest single population are Cypriots - which can be modelled as 63% Ancient Aegean!

You are taking these artificial labels too literally. The ancient Aegean was simply a part of the genetic eastern Mediterranean region in a way modern Greece is not. Cypriots are roughly halfway between Lebanese and Sicilians genetically because they have more Levantine ancestry than them but still share ancient components with southern Europe. It is likely this ancestry is, again, due to the "Sardinian-like" admixture found in all of southern Europe in the Sarno et al study that arrived from ancient Anatolia, not from Greece. There is no particular link between Cyprus and mainland Greece even the regions with lower Slavic ancestry, and the same is true for Lebanon. What you are seeing is ancient shared ancestry between all of the Mediterranean region due to a migration out of Anatolia, but you're labeling it "Aegean" when the Aegean was absolutely not the origin of this wave of migration, they were a recipient of it.


Later on, your best Oracle fit is Samaritan (66%) + Turkish (33%) - and modern Turks are descendants of Hellenized Byazntine Anatolians, an area which has been heavily colonized by Greeks since the 10th century BC all the way to the 15th century AD. You actually proved the model I've shown for them:

The spread of the Greek language in Anatolia was primarily a cultural, not a demographic shift, and most of the Greek population of Turkey were not absorbed into the Turkish population, they were expelled to Greece. Some of these also are genetically identical to their Anatolian neighbors (see Pontic Greeks, who are identical to Armenians in all of their components and clearly indigenous to the Caucasus region). So no, scoring Turkish on there is not a sign of ancient Greek admixture in Levantines, you're really reaching with that.


Agreed, and I've stated it before - Greek Islanders are much more Levantine mixed than mainland Greece. Could be for several reasons, but IMO it's either simple distance made the Aegeans closer to the Levant during Hellenistic times (when most admixture with Levantines occurred) or simply because the large Slavic influx in mainland Greece "pushed" down the admixture.

Unlikely that mainland Greeks ever had as much Levantine admixture as Sicilians or Aegean islanders. If you look at the bar chart from the Sarno study, look at the red component in mainland Greece... even if you took out all of the Slavic-related ancestry (blue component) and replaced it with red, they would still have less of the Near Eastern component than some regions of southern Italy or Cyprus. And if you removed the Slavic ancestry, the other components, not just Near Eastern, would also increase proportionally. Aegean islanders simply absorbed a recent wave of Levantine migration that never reached mainland Greece.



I have yet to examine Copts, so I would have to take your word for it. However, I'd find it very hard to believe, given the extensive colonization and the Greek city of Alexandria.

Again, Greeks did not settle en masse, it was mostly a cultural change, where those who adopted Greek language and culture identified as such. They were Hellenized natives, not transplants from Greece except for the ruling class.



Druze, Lebanese and even Palestinian Christians seem to have some Aegean admixture (20-30%).

You are mislabeling admixture as Aegean that really should be considered Anatolian that is simply SHARED between Levantines, Aegeans, Sicilians and Jews. The origin of the admixture was not in Greece.


I've never said they are - they are just more isolated than mainland Greece, especially when it comes to Slavic admixture.

They might not have received Slavic that did not come indirectly by way of the mainland but they did absorb Levantine and Anatolian Greek admixture so no, they were not more isolated overall, they just received different admixture.

kingjohn
06-08-2018, 12:01 PM
i look at my father ftdna family finder match
and he have a matching segments of 10cm with a non -jew Portuguese professor from porto
he didn't pass the segment to me bummer :(
the surname of this Portuguese individulal is originally spanish

p.s
my father score 39% sefhardi in my origins 2.0 so he is only partly ashkenazi

giorgio
06-08-2018, 12:14 PM
The issue is how you are labeling ancestral components, it is causing you to draw faulty conclusions.



You said "Aegean" but again there is no proof that this is the case nor, as has been demonstrated, is there an "Aegean" genetic profile. Because clearly as we see, Greek mainlanders are one genetic cluster, Aegean islanders are in the southern Italian cluster, and modern day Turks and Cypriots are distinct from both. It can also not be proven conclusively that in ancient times, even before Slavs and other groups migrated to the area that there were not also internal differences between mainland and islands.


Not all Greek Mainlanders plot in the same cluster. A Northern Greek would cluster closer to Bulgarians or Fyromanians than to most Peloponnese and especially compared to those from Laconia. Some Islanders pull towards the Mainland (Ionian, Western Aegean) while Rhodians and some Cretans get closer to Cypriots and Anatolian Greeks even more so than to Sicilians.




You are taking these artificial labels too literally. The ancient Aegean was simply a part of the genetic eastern Mediterranean region in a way modern Greece is not. Cypriots are roughly halfway between Lebanese and Sicilians genetically because they have more Levantine ancestry than them but still share ancient components with southern Europe. It is likely this ancestry is, again, due to the "Sardinian-like" admixture found in all of southern Europe in the Sarno et al study that arrived from ancient Anatolia, not from Greece. There is no particular link between Cyprus and mainland Greece even the regions with lower Slavic ancestry, and the same is true for Lebanon. What you are seeing is ancient shared ancestry between all of the Mediterranean region due to a migration out of Anatolia, but you're labeling it "Aegean" when the Aegean was absolutely not the origin of this wave of migration, they were a recipient of it.


The Sardinian like ancestry is rather mislabeled and represent an Early European Farmer migration from Anatolia which has spread across the Mediterranean sea and reached the Western/Atlantic parts of Europe. Even Northern Europeans have like 20-30% Neolithic Farmer usually labelled as "Sardinian" like admixture

There's a genetic link across the whole Southern European-Mediterranean Islander region. Slavic ancestry could be just general Steppe like geneflow which could've come from both Eastern Europe and Western Asia given Steppe like populations has spread in both regions. Mediterranean islanders just happen to receive less because they were more Isolated due to geographical reasons (primal example is Sardinia) Even Sicilians have like 10-15% Steppe like admixture, even Lebanese have a few percent.




The spread of the Greek language in Anatolia was primarily a cultural, not a demographic shift, and most of the Greek population of Turkey were not absorbed into the Turkish population, they were expelled to Greece. Some of these also are genetically identical to their Anatolian neighbors (see Pontic Greeks, who are identical to Armenians in all of their components and clearly indigenous to the Caucasus region). So no, scoring Turkish on there is not a sign of ancient Greek admixture in Levantines, you're really reaching with that.


I think we're bringing way too much facts here without being proven historically or genetically. Some Turks absorbed Greek speaking populations and certainly in Western Anatolia there's a high number of people with Greek ancestry, even with obvious "Turkified" Greek surnames. While genetically I don't think we can make a clear conclusion on what is Greek or not because we don't have actual proof on what populations formed the ethnogenesis of early Hellenes. Modern Greeks /= Hellenes altough they owe a good part of their ancestry to Hellenic like populations ~So do Sicilians, Albanians, Bulgarians Cypriots and some Turks.




Unlikely that mainland Greeks ever had as much Levantine admixture as Sicilians or Aegean islanders. If you look at the bar chart from the Sarno study, look at the red component in mainland Greece... even if you took out all of the Slavic-related ancestry (blue component) and replaced it with red, they would still have less of the Near Eastern component than some regions of southern Italy or Cyprus. And if you removed the Slavic ancestry, the other components, not just Near Eastern, would also increase proportionally. Aegean islanders simply absorbed a recent wave of Levantine migration that never reached mainland Greece.


I don't think Sicilians or Aegean Islanders have any significant Levantine admixture that should be worth to mention. If anything they have a good chump of Anatolian EBA like admixture which is also the dominant genetic component in the Levant (more so in the Northern part) than they share some direct Canaanite link - that would be quite unlikely. We're also talking about 5-10% differences here so making an assumption based on a single DNA study isn't much of a proof or too weak to bring any serious conclusions here.
If anything I think the majority of the Near Eastern like ancestry in South Europe is rather more Native than the recent Steppe, Indo European like geneflow which dramatically changed the genetic structure of Mainland Greece. I don't think Steppe people were European or any native rather a group of people who lived in the Eastern steppes and had absorbed many different ethnic groups. Some of them could've been Iranian like while others were more Estonian like (just an example of their genetic cline)

Claudio
06-08-2018, 12:20 PM
The issue is how you are labeling ancestral components, it is causing you to draw faulty conclusions.



You said "Aegean" but again there is no proof that this is the case nor, as has been demonstrated, is there an "Aegean" genetic profile. Because clearly as we see, Greek mainlanders are one genetic cluster, Aegean islanders are in the southern Italian cluster, and modern day Turks and Cypriots are distinct from both. It can also not be proven conclusively that in ancient times, even before Slavs and other groups migrated to the area that there were not also internal differences between mainland and islands.





Or it could be the other way around. Or both groups could simply have older customs because they are earlier established communities that remained isolated and possibly in contact with one another. Again, we do not know.




I do think that Phoenician admixture explains some of what is going on as well as other groups that were in Sicily like Elymians (relatives of the Hittites who would have carried high Caucasus type ancestry), but I think that you still see some degree of Near Eastern admixture in places where there was no historical Phoenician nor Carthaginian settlement shows that some of it is more ancient than that.




That study stated that some of the admixture is recent, but it didn't state that all of it is.



It could be an accumulation of multiple admixture events but some of it is likely to be more ancient than that.






We also do not know for a fact that result represents all of Sicily's population in its era. The Beaker culture came to Sicily from Sardinia and the Sarno et al study you have quoted states that Trapani (which is where Partanna is located) differs from the rest of Sicily due to two factors: higher East European admixture (which could be Greek) and higher Sardinian-like admixture which is recent. This could be the Beakers. If you took a sample from the same era from a different part of Sicily you would likely find something resembling Minoans considering that the sample from Partanna was basically similar to Minoans with extra Sardinian admixture. But within that they still had SOME degree of Near Eastern ancestry, showing it is not entirely from Phoenicians and Arabs even if some of it definitely is. Modern Sicilians are more Near Eastern than that sample, but some of that ancestry did already exist there.




Considering that most of Sicily and Calabria have no IBD sharing with mainland Greece and very low NE European admixture (unlike Apulia which has higher IBD sharing with Greece and Eastern Europe -- see the Sarno study again), it is unlikely there was a recent mass migration of mainland Greeks to Calabria. In fact of all southern Italians, the people of southern Calabria have the oldest DNA and they have mixed very little with non-locals. The Greek speaking population especially, but they still do not show IBD sharing with modern Greeks which might suggest that they acquired the Greek language very early on and then did not see any recent infusion of foreign DNA.




The suggestion is that the non-Levantine DNA in Ashkenazim is from northern Italy. Northern Italians share some degree of ancestry with French and Germans, especially considering they have both Gallic and Germanic ancestry.




You are taking these artificial labels too literally. The ancient Aegean was simply a part of the genetic eastern Mediterranean region in a way modern Greece is not. Cypriots are roughly halfway between Lebanese and Sicilians genetically because they have more Levantine ancestry than them but still share ancient components with southern Europe. It is likely this ancestry is, again, due to the "Sardinian-like" admixture found in all of southern Europe in the Sarno et al study that arrived from ancient Anatolia, not from Greece. There is no particular link between Cyprus and mainland Greece even the regions with lower Slavic ancestry, and the same is true for Lebanon. What you are seeing is ancient shared ancestry between all of the Mediterranean region due to a migration out of Anatolia, but you're labeling it "Aegean" when the Aegean was absolutely not the origin of this wave of migration, they were a recipient of it.



The spread of the Greek language in Anatolia was primarily a cultural, not a demographic shift, and most of the Greek population of Turkey were not absorbed into the Turkish population, they were expelled to Greece. Some of these also are genetically identical to their Anatolian neighbors (see Pontic Greeks, who are identical to Armenians in all of their components and clearly indigenous to the Caucasus region). So no, scoring Turkish on there is not a sign of ancient Greek admixture in Levantines, you're really reaching with that.



Unlikely that mainland Greeks ever had as much Levantine admixture as Sicilians or Aegean islanders. If you look at the bar chart from the Sarno study, look at the red component in mainland Greece... even if you took out all of the Slavic-related ancestry (blue component) and replaced it with red, they would still have less of the Near Eastern component than some regions of southern Italy or Cyprus. And if you removed the Slavic ancestry, the other components, not just Near Eastern, would also increase proportionally. Aegean islanders simply absorbed a recent wave of Levantine migration that never reached mainland Greece.




Again, Greeks did not settle en masse, it was mostly a cultural change, where those who adopted Greek language and culture identified as such. They were Hellenized natives, not transplants from Greece except for the ruling class.




You are mislabeling admixture as Aegean that really should be considered Anatolian that is simply SHARED between Levantines, Aegeans, Sicilians and Jews. The origin of the admixture was not in Greece.



They might not have received Slavic that did not come indirectly by way of the mainland but they did absorb Levantine and Anatolian Greek admixture so no, they were not more isolated overall, they just received different admixture.

I think I have to agree with Sikeliot on this occasion.
Although I disagree with his opinion of the people from the tip of Calabria.
Though Calabrians are largely Ancient Greek in Ancestry as far as Levantine ancestry goes Calabria was invaded numerous time by Arabs and also occupied with settlements and farming introduction of lemons & oranges etc..
and the population of Reggio was even sold into slavery.
Plus Calabria has had a long and ancient Jewish history to the point in the Middle Ages that In some places in Calabria the populations were recorded as being up to 50% Jewish.
So I would imagine those things are the cause of elevated Levantine in Calabrian's not ancient Levantine Admixture.
Come to think of it on a similar note Sicily’s population actually doubled under Muslim Occupation to the point that Palermo even became the largest city in the Arab world after Baghdad.

giorgio
06-08-2018, 12:27 PM
I think I have to agree with Sikeliot on this occasion.
Although I disagree with his opinion of the people from the tip of Calabria.
Though Calabrians are largely Ancient Greek in Ancestry as far as Levantine ancestry goes Calabria was invaded numerous time by Arabs and also occupied with settlements and farming introduction of lemons & oranges etc..
and the population of Reggio was even sold into slavery.
Plus Calabria has had a long and ancient Jewish history to the point in the Middle Ages that In some places in Calabria the populations were recorded as being up to 50% Jewish.
So I would imagine those things are the cause of elevated Levantine in Calabrian's not ancient Levantine Admixture.

Can you post some link on the various Arab occupation in Calabria? I can't find a single historical article about it. Thanks.

Claudio
06-08-2018, 12:43 PM
Can you post some link on the various Arab occupation in Calabria? I can't find a single historical article about it. Thanks.

Wikipedia for starters..23791

Though I would say Calabria’s Jewish history is probably much more responsible for elevated Levantine than any Arab Muslim Ancestry.

giorgio
06-08-2018, 01:02 PM
Wikipedia for starters..23791

Though I would say Calabria’s Jewish history is probably much more responsible for elevated Levantine than any Arab Muslim Ancestry.

I think generally Calabrians are more Byzantine (higher East Med with less SSA and North African and Arabian) where as Sicilians are more North African, Arabian and less East Mediterranean with slightly more North European (probably North Italian settlements) based on Eurogenes.

Erikl86
06-08-2018, 01:18 PM
I think you assume a lot of stuff, plus you seem to hand wave a lot of historical facts and scientific results, in order to justify your arguments.






You said "Aegean" but again there is no proof that this is the case nor, as has been demonstrated, is there an "Aegean" genetic profile. Because clearly as we see, Greek mainlanders are one genetic cluster, Aegean islanders are in the southern Italian cluster, and modern day Turks and Cypriots are distinct from both. It can also not be proven conclusively that in ancient times, even before Slavs and other groups migrated to the area that there were not also internal differences between mainland and islands.


See, this is where you are wrong. Cypriots cluster with Island Greeks and S. Italians and Sicilians. They form, at least today, something Lazirdis call "Mediterranean continuum". They are distinct somewhat from mainland Greeks and West Anatolian Turks, but only to a degree (East and Central Anatolian Turks are something else).

The Graph I've provided, clearly show that Cypriots are Aegean heavily mixed with Levantines. Also, historically, they have been settled by the Mycenaeans.


Or it could be the other way around. Or both groups could simply have older customs because they are earlier established communities that remained isolated and possibly in contact with one another. Again, we do not know.

No. Romanoites are older - even BC. Also, since Judaism started off in the Levant, logic dictates Jews moved from Eastern Mediterranean to the West. This is historically and geographically wrong assumption.



I do think that Phoenician admixture explains some of what is going on as well as other groups that were in Sicily like Elymians (relatives of the Hittites who would have carried high Caucasus type ancestry), but I think that you still see some degree of Near Eastern admixture in places where there was no historical Phoenician nor Carthaginian settlement shows that some of it is more ancient than that.


People, in the last 3000 years, have been moving around. Or you assume that if Phoenician colonization was more concentrated in Western Sicily, that they just stayed there? This is absurd. And to conclude from there, that it must be older because Phoenicians didn't colonize that specific village, but 10-20 km away some 3000 years ago - that is just bad logic.


That study stated that some of the admixture is recent, but it didn't state that all of it is.
The study clearly shown that Near Eastern admixture among Greek Islanders (beside Cyprus, which was a Phoenician colony) is post Mycenaean - the oldest one is estimated to be 3800 YBP (years before present) - the start of the First Wave of Greek Colonization. The rest is dated after the 3th century BC and well into the first millenium AD, which would make sense because the Levant has been a conflict zone between Zoroastrian Persian Empires and Christian Byzantine Empire for centuries, probably driving a lot of Christian immigrant Westward.



It could be an accumulation of multiple admixture events but some of it is likely to be more ancient than that.

More assumptions - that actually does not get a long with history or genetic evidence. Minoans and Mycenaeans show no Near Eastern admixture - as I have shown. Only Mycenaeans show some - which was communicated via BA Anatolian immigration (which Mycenaean were 6% admixed).





We also do not know for a fact that result represents all of Sicily's population in its era. The Beaker culture came to Sicily from Sardinia and the Sarno et al study you have quoted states that Trapani (which is where Partanna is located) differs from the rest of Sicily due to two factors: higher East European admixture (which could be Greek) and higher Sardinian-like admixture which is recent. This could be the Beakers. If you took a sample from the same era from a different part of Sicily you would likely find something resembling Minoans considering that the sample from Partanna was basically similar to Minoans with extra Sardinian admixture. But within that they still had SOME degree of Near Eastern ancestry, showing it is not entirely from Phoenicians and Arabs even if some of it definitely is. Modern Sicilians are more Near Eastern than that sample, but some of that ancestry did already exist there.


It might not represent all of Sicily - but this is what we need to work with. Also, again - you assume there would be more ancient significant Near Eastern admixture - without any base what so ever. I employ simple logic here - Sicily and S. Italy have been a destination for mass immigration and colonization of Greeks and Near Eastern people - and it is well documented. Seeing the qpAdm models for modern Sicilians vs. BA Sicilians - modern Sicilians today are only about 20% BA Sicilians. The rest is Greek and Near Eastern.



Considering that most of Sicily and Calabria have no IBD sharing with mainland Greece and very low NE European admixture (unlike Apulia which has higher IBD sharing with Greece and Eastern Europe -- see the Sarno study again), it is unlikely there was a recent mass migration of mainland Greeks to Calabria. In fact of all southern Italians, the people of southern Calabria have the oldest DNA and they have mixed very little with non-locals. The Greek speaking population especially, but they still do not show IBD sharing with modern Greeks which might suggest that they acquired the Greek language very early on and then did not see any recent infusion of foreign DNA.
I have never said MASS MIGRATION, I said, and gave historical reference to it, that such movement had occured. Obviously, the fact that S. Italians and Sicilians cluster much closer with Greek Islanders than mainland Greeks, prove it wasn't massive. Also, because when this migration occurd S. Italy and Sicily have already been deep in the process of "Italiazion" it might be that those mainland Greeks didn't settle in locations that genetically were highly Greek, but linguistically not anymore.




You are taking these artificial labels too literally. The ancient Aegean was simply a part of the genetic eastern Mediterranean region in a way modern Greece is not. Cypriots are roughly halfway between Lebanese and Sicilians genetically because they have more Levantine ancestry than them but still share ancient components with southern Europe. It is likely this ancestry is, again, due to the "Sardinian-like" admixture found in all of southern Europe in the Sarno et al study that arrived from ancient Anatolia, not from Greece. There is no particular link between Cyprus and mainland Greece even the regions with lower Slavic ancestry, and the same is true for Lebanon. What you are seeing is ancient shared ancestry between all of the Mediterranean region due to a migration out of Anatolia, but you're labeling it "Aegean" when the Aegean was absolutely not the origin of this wave of migration, they were a recipient of it.

You contradict yourself here - earlier you claimed Cypriots are distinct and do not cluster with Sicilians and Greek Islanders etc..
Anyhow, you are almost right - the Mediterranean, ancient genetic make up was Sardinian-like. Minoans were also - Sardinian-like. Levantines, were not and formed distinct genetic component which is very easy to tell apart and also comes up really easily when you look at admixtures.

Also, you are over exaggerating the distinctiveness of mainland Greeks - look at the graph I've provided from that study - the average mainland Greek is maybe 20-25% admixed with Slavs. The rest is pretty consistent with Island Greeks, Cypriots, Sicilians and S. Italians.



The spread of the Greek language in Anatolia was primarily a cultural, not a demographic shift, and most of the Greek population of Turkey were not absorbed into the Turkish population, they were expelled to Greece. Some of these also are genetically identical to their Anatolian neighbors (see Pontic Greeks, who are identical to Armenians in all of their components and clearly indigenous to the Caucasus region). So no, scoring Turkish on there is not a sign of ancient Greek admixture in Levantines, you're really reaching with that.

Not so. In the majority of cases, perhaps. But there was major wave of Greek immigration and colonization. It's also funny you are open minded to the fact that whole of S. Italy and Sicily were colonized with Greeks, to the point that genetically they are closer to Greeks than Northern Italians, but you seem to reject the simple historical fact that just next door - to Western Anatolia and much closer Northern Egypt - the didn't. I've brought in this thread quote about mass immigration of classical and Hellenistic Greeks from Greece mainland and Islands to the Levant.

Also, Western Anatolian Turks are very close genetically to Greeks. So, it depends where from in Turkey. The two populations also share very similar Y-Haplogroups types and proportions. Central Turks are more Anatolian, and Eastern Turks are more Armenian and Caucasian. Southern Turks are heavily Iranian, Anatolian and even a little bit Levantine.


Unlikely that mainland Greeks ever had as much Levantine admixture as Sicilians or Aegean islanders.
Where on earth did I ever say that? If ever - I specifically remember me saying they have LESS Levantine admixture.


If you look at the bar chart from the Sarno study, look at the red component in mainland Greece... even if you took out all of the Slavic-related ancestry (blue component) and replaced it with red, they would still have less of the Near Eastern component than some regions of southern Italy or Cyprus. And if you removed the Slavic ancestry, the other components, not just Near Eastern, would also increase proportionally. Aegean islanders simply absorbed a recent wave of Levantine migration that never reached mainland Greece.

This is just repeating what I said and what I brought. Also, obviously there was Levantine migration into mainland Greece, or the red bar for them would be like the blue bar (representing North/Eastern European-like admixture) for Cypriots.



Again, Greeks did not settle en masse, it was mostly a cultural change, where those who adopted Greek language and culture identified as such. They were Hellenized natives, not transplants from Greece except for the ruling class.
Depends where, but in some places they did, and you have genetic evidence for it. For god sake - Sicily and S. Italy are an excellent example of that.




You are mislabeling admixture as Aegean that really should be considered Anatolian that is simply SHARED between Levantines, Aegeans, Sicilians and Jews. The origin of the admixture was not in Greece.
The distinction is not so sharp, but quite blurry. Read about ancient Anatolian Greek city states, like Troy, or Antioch.



They might not have received Slavic that did not come indirectly by way of the mainland but they did absorb Levantine and Anatolian Greek admixture so no, they were not more isolated overall, they just received different admixture.
Agreed.

Sikeliot
06-08-2018, 01:35 PM
I think generally Calabrians are more Byzantine (higher East Med with less SSA and North African and Arabian) where as Sicilians are more North African, Arabian and less East Mediterranean with slightly more North European (probably North Italian settlements) based on Eurogenes.

I notice Dodecanese and Calabria shifting toward Cyprus but Sicily shifting toward North African Jews, at their extremes.

Erikl86
06-08-2018, 01:36 PM
So many remarks...

I cannot imagine that Greek islanders hsving real Levantine ancestry, what they have is the Anatolian Farmer component which unites us all (I think that was Sikeliot's point?). As far as I can see some of you guys are mixing up statistical significance with historical reality. Life is not so easy, at least not for me. Whatever mathematical method we use, as a molecular biologist I can see that the data is derived from SNPs and assigned to populations. However most SNPs are common among historically recent populations. For example both my parents show around 40% European farmer component in "Kurd"s K14 geneplaza test. One of the highests among the tested people so far. But as far as I can see, he uses the distinguishing SNPs instead. Even if the european jews were descendants from Greeks still most Jews lived in Anatolia and Thrace at that time and much less in Greece and in the islands proper. How close were the Hellenized populations to the Greeks genetically?

Still when i saw the G25 PCA of Davidski, i was really surprised that both Ashkenazi and Sefardic samples appearso close to Eastern med. and distinct from Iraqi Jews or Samaritans. I would have expected Ashkenazi more shifted to the north due to supposed admixture and Sefardi closer to Spain or to North African Jews etc. In fact they look more southern in the PCA (eastern in reality) than the Italians. This cannot be a coincidence. The Greeks are not there alone but also Anatolians and Lebanese Christians are closeby etc. The Sea people might have come from the islands, or cilicia or cyprus. In any case they mixed very soon with the local population.

Finally How do samaritans represent an ancient native population? Was there no conversion to Samartian religion, for me they are a relic of something that included many converts and settlers from Assyria.

There are multiple scenarios that could be statistically fit and logical but difficult to know which one(s) are real.


23788

I agree with your assumptions - I'm also not claiming that there is no Levantine ancestry to Western Jews. On the contrary - I claim it's about 30%-35%. Also, most likely most of the converts were women - as evidenced by historical sources (circumcision for grown men would deter many, plus circumcised genitalia was considered ugly in the classical world), because Western Jews seem to share Y Haplogroups with Levantines - same types and similar proportions:

Lebanese Y Haplogroups:
23794

Ashkenazi Y Haplogroups:
23795

And Eurogenes K36 similarity maps for Sephadic Jews average, and Ashkenazi Jews average. Guess who's who?

23796

23797

Michalis Moriopoulos
06-08-2018, 01:47 PM
It's exciting to think just how informative a mere handful of Iron Age genomes from the Mezzogiorno, Aegean, and Levant would be for so many modern populations. The most burning questions we have about East Mediterranean demography could conceivably be solved in just one paper. What a time to be an anthropologist...

kingjohn
06-08-2018, 01:57 PM
It's exciting to think just how informative a mere handful of Iron Age genomes from the Mezzogiorno, Aegean, and Levant would be for so many modern populations. The most burning questions we have about East Mediterranean demography could conceivably be solved in just one paper. What a time to be an anthropologist...

the consequence of such paper could be a disaster to sikeliot thinking .... :)
regards
adam

Sikeliot
06-08-2018, 02:14 PM
the consequence of such paper could be a disaster to sikeliot thinking .... :)
regards
adam

No, it would be a disaster to the idea of there being only limited Slavic input in Greece. It would show as the Mycenaean sample does that there is a large Slavic element there.

Sikeliot
06-08-2018, 02:26 PM
I agree with your assumptions - I'm also not claiming that there is no Levantine ancestry to Western Jews. On the contrary - I claim it's about 30%-35%. Also, most likely most of the converts were women - as evidenced by historical sources (circumcision for grown men would deter many, plus circumcised genitalia was considered ugly in the classical world), because Western Jews seem to share Y Haplogroups with Levantines - same types and similar proportions:

Lebanese Y Haplogroups:
23794

Ashkenazi Y Haplogroups:
23795

And Eurogenes K36 similarity maps for Sephadic Jews average, and Ashkenazi Jews average. Guess who's who?

23796

23797

We agree on the genetic reality today but we just disagree slightly on why it is that way. I definitely am not denying Phoenician and Arab ancestry in Sicily, I'm proud of it in fact. I just think there's more to the story also is all!

Jews may have a partial Arabian ancestry from ancient times going by autosomal and their haplogroup J frequency.

giorgio
06-08-2018, 02:32 PM
I notice Dodecanese and Calabria shifting toward Cyprus but Sicily shifting toward North African Jews, at their extremes.

Calabrians, Sicilians are primarily Barchin Neolithic or Anatolian EBA with Italic and secondary input from other areas. ie. Aberesche in some regions or Arabic/Maghrebi in some.
Dodecanese are the most Eastern shifted region of Greece (geography speaking would be part of Anatolia rather Europe) so their gedmatch scores aren't that surprising.. Calabria on the other hand is more outlying for their region expecting them to be more Western shifted than Greek Islanders but a lot are just as Near East shifted as the most outlying Greek Islands.

For example a Dodecanese Greek from Karpathos gets very similar to South Italians but have East European admixture instead of the Western European which captures the Continental European ancestry in Sicilians.

# Population Percent
1 SW_Europe 34.13
2 West_Asia 27.62
3 SW_Asia 20.57
4 NE_Europe 13.58
5 Siberia 1.08
6 South_Asia 1.08
7 NE_Asia 0.9
8 West_Africa 0.69
9 Oceania 0.34

Single Population Sharing:

# Population (source) Distance
1 Sephardic_Jew 4.86
2 Ashkenazy_Jew 7.12
3 Italian_Sicilian 7.42
4 Cypriot 9.07
5 Turkish 10.56
6 Greek_Central 11.19
7 Italian_Abruzzo 11.91
8 Syrian 12.09
9 Lebanese_Christian 12.28
10 Turkish_Kayseri 12.34
11 Lebanese_Druze 12.83
12 Lebanese_Muslim 13.61
13 Jordanian 13.76
14 Turkish_Aydin 14.2
15 Palestinian 16.4
16 Greek_Thessaly 16.65
17 Albanian 17.26
18 Azerbaijan_Azeri 18.1
19 Assyrian 18.34
20 Samaritan_Jew 18.35

Mixed Mode Population Sharing:

# Primary Population (source) Secondary Population (source) Distance
1 69.9% Sephardic_Jew + 30.1% Turkish @ 1.91
2 69.4% Lebanese_Muslim + 30.6% Spaniard @ 1.96
3 73.3% Sephardic_Jew + 26.7% Turkish_Kayseri @ 1.96
4 76.1% Sephardic_Jew + 23.9% Turkish_Aydin @ 2
5 51.6% Albanian + 48.4% Samaritan_Jew @ 2.06
6 70.7% Lebanese_Druze + 29.3% Spaniard @ 2.15
7 52.5% Greek_Thessaly + 47.5% Samaritan_Jew @ 2.16
8 71.7% Lebanese_Christian + 28.3% Spaniard @ 2.2
9 66.8% Ashkenazy_Jew + 33.2% Jordanian @ 2.25
10 70.6% Ashkenazy_Jew + 29.4% Palestinian @ 2.26
11 80.7% Sephardic_Jew + 19.3% Azerbaijan_Azeri @ 2.29
12 84.7% Sephardic_Jew + 15.3% Iranian @ 2.3
13 62.5% Greek_Central + 37.5% Samaritan_Jew @ 2.36
14 74.3% Lebanese_Christian + 25.7% French @ 2.38
15 67.6% Lebanese_Christian + 32.4% Italian_Bergamo @ 2.39
16 61.2% Lebanese_Christian + 38.8% Italian_Tuscan @ 2.4
17 84.5% Sephardic_Jew + 15.5% Dagestan_Azeri @ 2.46
18 50.1% Samaritan_Jew + 49.9% Kosovar @ 2.48
19 83.8% Sephardic_Jew + 16.2% Kurdish @ 2.49
20 58.7% Lebanese_Christian + 41.3% Albanian @ 2.5


South Aegean didn't receive any noticeable Jewish or Levantine input still look at how East Mediterranean shifted and this is likely to be Anatolia EBA rather direct Levant ancestry. This sample is basically zero Mainland Greek like and would fall in the Anatolian Greek cluster as expected for East Aegean heavily colonized by Ionian Greeks and lack the Steppe, Balkanic input.

Targum
06-08-2018, 02:35 PM
As to the chronology of the two "seed" communities of Western Jews; Romaniote and Italqim. These are the two original Jewish communities in Europe. They are distinct, but both result from the original migration of Jews from Israel to the Greek-speaking world of the time. When Jews first came to Greece, the Italic identity had not yet become "Roman". The Jews in Rome proper, developed this identity as Italqim via the Roman host culture. The Romaniote developed a Judeo-Greek culture and language (Yevanic-Judeo-Greek written in Hebrew, like Yiddish, Ladino and Yemenite Judeo-Arabic). So technically the Romaniote were in place earlier, even though the Romaniote distinctiveness also developed over history and was subject to the evolutionary influences of being under the Christian-Byzantine avatar of Greek culture, obviously not the same as the Classical Greek world to which the Jews first came. Once Italqim and the Roman culture were in place, these two communities developed separately, though like all Jews remained in contact. The later ethnogenesis of Ashkenazi, Sefaradi, and Syrian Jews( Western Jews grafted onto musta'arib Jews in place from Israel) was an outgrowth of both of them and both are the ancestors of Western Jews.

kingjohn
06-08-2018, 02:51 PM
We agree on the genetic reality today but we just disagree slightly on why it is that way. I definitely am not denying Phoenician and Arab ancestry in Sicily, I'm proud of it in fact. I just think there's more to the story also is all!

Jews may have a partial Arabian ancestry from ancient times going by autosomal and their haplogroup J frequency.



no ;)
we have levantine ancestery not arabian
i score 0% arabian in livingdna but i do score significant levantine
same goes for johns who is full aschenazi he score only 1-1.5% arabia in living dna extremely low .....

Claudio
06-08-2018, 02:52 PM
I think generally Calabrians are more Byzantine (higher East Med with less SSA and North African and Arabian) where as Sicilians are more North African, Arabian and less East Mediterranean with slightly more North European (probably North Italian settlements) based on Eurogenes.

I was about to mention Byzantine influences.
If we look at the Byzantine Empire stretching past Greece,Turkey,Armenia,Levant,And down to Egypt.
Two of the main ethnic groups through out the Byzantine reign were Greek Speeking Armenians & Syrians.
When Justian re-conquered italy he used a Greco-Armenian/Syrian Army.
It’s also documented he re-populated Southern Italy with amongst others Armenian populations.
Calabria was the Southern capital of Byzantine Italy and had the longest running influence from These Byzantine populations demographics influence so to Speak,this coupled with what I mentioned of Calabria’s large and long running Jewish history.
So Calabrians scoring high levantine & Caucasus is probably mainly as a result of having a bigger combination of Byzantine & Jewish Admixture rather than than just Arab Muslim,mixed with its Ancient Greek Italian population.
The opposite of a ancient genetic isolate population.
With admix left right and center just like the rest of Southern Italy. ;)

Sikeliot
06-08-2018, 03:29 PM
I was under the impression Calabria was genetically isolated but maybe I am wrong. The only thing I know is that the low IBD with Greece today means the Greek speakers there likely did not come from the mainland. Maybe they came from Anatolia. Because the regions with higher IBD with mainland Greece also show elevated East Euro autosomal input.

giorgio
06-08-2018, 05:30 PM
no ;)
we have levantine ancestery not arabian
i score 0% arabian in livingdna but i do score significant levantine
same goes for johns who is full aschenazi he score only 1-1.5% arabia in living dna extremely low .....

Levantines and Arabians overlap and often indistinguishable on DNA maps. It's certainly obvious that without the religious attributes both Jews and Arabs share common ancestry with some differences. Take example the difference between Christian Palestinians and Gaza Palestinians - On DNA they would both appear similar despite the Gazans have higher affinity with Egyptians but the vast majority of their DNA would be the same. I think the Original Hebrews (assuming they were homogeneous at the beginning) were similar to Canaanites or let's say modern Samaritans. Eastern Hebrews like Iraqi Jews or Iranian Jews might be descendants of converted Babylonians or Assyrians or intermarried with those groups, where as Western Jews mixed with other East Mediterranean groups (Myceaneans, Romans, Anatolians to some extent and later some minimal admix from Central and North Europeans.

Psynome
06-08-2018, 05:37 PM
It's exciting to think just how informative a mere handful of Iron Age genomes from the Mezzogiorno, Aegean, and Levant would be for so many modern populations. The most burning questions we have about East Mediterranean demography could conceivably be solved in just one paper. What a time to be an anthropologist...

https://www.archaeoscience.org

"DNA is now beginning to illuminate the period that saw the rise of civilizations in the ancient Mediterranean. Accordingly, the Initiative for the Science of the Human Past at Harvard (SoHP) is delighted to announce the formation of a new center for the study of the Mediterranean using ancient DNA and other scientific approaches.
The Max Planck-Harvard Research Center for the Archaeoscience of the Ancient Mediterranean (MHAAM) is a platform to engage colleagues and students in the discovery of new data which will prompt us to re-think and revise many of our contemporary perspectives on the history of pandemic disease, cultural engagement, migration and human health."


In the next few years, I'm hoping this effort will give us the answers we're looking for. I don't know of any papers published by them yet, but clearly there are big plans in store.

kingjohn
06-08-2018, 05:41 PM
Levantines and Arabians overlap and often indistinguishable on DNA maps. It's certainly obvious that without the religious attributes both Jews and Arabs share common ancestry with some differences. Take example the difference between Christian Palestinians and Gaza Palestinians - On DNA they would both appear similar despite the Gazans have higher affinity with Egyptians but the vast majority of their DNA would be the same. I think the Original Hebrews (assuming they were homogeneous at the beginning) were similar to Canaanites or let's say modern Samaritans. Eastern Hebrews like Iraqi Jews or Iranian Jews might be descendants of converted Babylonians or Assyrians or intermarried with those groups, where as Western Jews mixed with other East Mediterranean groups (Myceaneans, Romans, Anatolians to some extent and later some minimal admix from Central and North Europeans.


yes i score indo -iranian in dna land and gencove and also pakistan in dna tribes snp tes
and south central asia in k29 calculator of kurd
yes mizrachi jews acquired some aryan genes during the 2000 years in the persian yard .:D

Erikl86
06-08-2018, 08:47 PM
I don't think Sicilians or Aegean Islanders have any significant Levantine admixture that should be worth to mention. If anything they have a good chump of Anatolian EBA like admixture which is also the dominant genetic component in the Levant (more so in the Northern part) than they share some direct Canaanite link - that would be quite unlikely. We're also talking about 5-10% differences here so making an assumption based on a single DNA study isn't much of a proof or too weak to bring any serious conclusions here.
If anything I think the majority of the Near Eastern like ancestry in South Europe is rather more Native than the recent Steppe, Indo European like geneflow which dramatically changed the genetic structure of Mainland Greece. I don't think Steppe people were European or any native rather a group of people who lived in the Eastern steppes and had absorbed many different ethnic groups. Some of them could've been Iranian like while others were more Estonian like (just an example of their genetic cline)

I mostly agree with you, until this part ^^^

For two main reasons:

1. Phoenician colonization and settlement in Sicily is well documented, as well as Greek. Sicily and South Italy at one point, had specific Greek dialects directly derive from Greece mainland and Aegean Islands (I'm talking 2000 years ago, yes?). Modern qpAdm modelling of Sicilians always include substantial Levantine admixture - about ~10% Levantine, ~10% North African. Around 37% of it is Greek - as in derived from those Greek colonialists.
2. While BA Anatolian exist to some degree among modern-day Levantines, it's minor, and the model many used today is a model suggested by Hebar et al and Lazirdis in the last few years of Neolithic Levantine + CHL Iranian, an admixture which occured around 6000 YBP, and kind of eliminates BA Anatolian admixture which might get overlap with Southern Europeans. This is also the model I use. I know we're already at page 11, but if you read the previous pages, you should see it. In any case, here are my Basal-rich K-7 models:


[1] "distance%=0.0292 / distance=0.000292"

Lebanese_Christian:average

Minoan_Lasithi:I0070 31.55
Iran_ChL:I1661 25.15
Levant_N:I1704 24.50
Iran_ChL:I1665 12.55
Mycenaean:I9041 4.30
Levant_N:I1699 1.65
Iran_ChL:I1662 0.30

Roughly 65% Bronze Age Levantines, 35% ancient Aegean.

Distance of 0.0292%.

Let's check Samaritans, with the exact same populations as Christian Lebanese:

[1] "distance%=0.5681 / distance=0.005681"

Samaritan:GSM537032

Iran_ChL:I1665 34.5
Levant_N:I1704 33.2
Levant_N:I0867 32.4

No Aegean admixture appeared.

Let's check Christian Israeli Arabs:

[1] "distance%=0.134 / distance=0.00134"

Arab_Israel_1:average

Levant_N:I1704 30.2
Minoan_Lasithi:I0073 26.6
Iran_ChL:I1665 21.8
Iran_ChL:I1662 15.9
Iran_ChL:I1661 5.5

Roughly 73% Bronze Age Levantines, 26% Aegean - mostly Minoan (No Mycenaean).

Next, let's check Druze:

[1] "distance%=0.0289 / distance=0.000289"

Druze:average

Minoan_Lasithi:I0073 21.20
Levant_N:I1704 20.35
Iran_ChL:I1665 19.65
Iran_ChL:I1661 14.40
Mycenaean:I9010 9.15
Iran_ChL:I1662 8.05
Mycenaean:I9006 6.50
Iran_ChL:I1674 0.70

Very similar to Christian Lebanese - Bronze Age Levantine is 63%, Ancient Aegean is 37% (both Minoan and Mycenaean). Distance is 0.0289%.

Now let's check Cypriots:

[1] "distance%=0.0998 / distance=0.000998"

Cypriot:average

Minoan_Odigitria:I9131 24.80
Mycenaean:I9041 21.40
Minoan_Lasithi:I0073 17.10
Iran_ChL:I1661 12.00
Iran_ChL:I1662 8.65
Iran_ChL:I1665 7.25
Levant_N:I1704 6.70
Iran_ChL:I1674 2.10

Mirror image of Druze . 63% Aegean, 37% Levantines. Not surprising at all.

BA Anatolian surely would infiltrate, especially via Hurrians, Hittites and Mitanni which were BA Anatolian people that established empires which have reigned over parts of the Levant from 21st to the 12th centuries BC. But, this N_Levant + CHL_Iranian model helps eliminate it and give us the possibility to locate Levantine admixture among present day Sicilians and Aegean Islanders.

Erikl86
06-08-2018, 08:49 PM
I was about to mention Byzantine influences.
If we look at the Byzantine Empire stretching past Greece,Turkey,Armenia,Levant,And down to Egypt.
Two of the main ethnic groups through out the Byzantine reign were Greek Speeking Armenians & Syrians.
When Justian re-conquered italy he used a Greco-Armenian/Syrian Army.
It’s also documented he re-populated Southern Italy with amongst others Armenian populations.
Calabria was the Southern capital of Byzantine Italy and had the longest running influence from These Byzantine populations demographics influence so to Speak,this coupled with what I mentioned of Calabria’s large and long running Jewish history.
So Calabrians scoring high levantine & Caucasus is probably mainly as a result of having a bigger combination of Byzantine & Jewish Admixture rather than than just Arab Muslim,mixed with its Ancient Greek Italian population.
The opposite of a ancient genetic isolate population.
With admix left right and center just like the rest of Southern Italy. ;)

I was unaware of Syrian Byzantines being so important in the Empire - I was aware of Armenian importance. That would actually explain why most Levantine admixture among Greeks is dated to Byzantine period - probably came from those Syrians.

Erikl86
06-08-2018, 08:57 PM
As to the chronology of the two "seed" communities of Western Jews; Romaniote and Italqim. These are the two original Jewish communities in Europe. They are distinct, but both result from the original migration of Jews from Israel to the Greek-speaking world of the time. When Jews first came to Greece, the Italic identity had not yet become "Roman". The Jews in Rome proper, developed this identity as Italqim via the Roman host culture. The Romaniote developed a Judeo-Greek culture and language (Yevanic-Judeo-Greek written in Hebrew, like Yiddish, Ladino and Yemenite Judeo-Arabic). So technically the Romaniote were in place earlier, even though the Romaniote distinctiveness also developed over history and was subject to the evolutionary influences of being under the Christian-Byzantine avatar of Greek culture, obviously not the same as the Classical Greek world to which the Jews first came. Once Italqim and the Roman culture were in place, these two communities developed separately, though like all Jews remained in contact. The later ethnogenesis of Ashkenazi, Sefaradi, and Syrian Jews( Western Jews grafted onto musta'arib Jews in place from Israel) was an outgrowth of both of them and both are the ancestors of Western Jews.

Yep. Also worth mentioning that there were Greek-speaking Romaniote community in S. Italy - and not the other way around (there was no documented Italkim communities in the Eastern Mediterranean). See: https://en.wikipedia.org/wiki/Italian_Jews#Graeco-Italian_Jews

Agamemnon
06-08-2018, 09:05 PM
Jews may have a partial Arabian ancestry from ancient times going by autosomal and their haplogroup J frequency.

There's only one branch of J1 among Ashkenazi Jews which can be securely labeled Arabian, namely ZS2102 which is a branch of FGC8712 exactly like L859 (the main Hashemite branch). This might actually be Nabatean in origin, if it isn't older that is.

Erikl86
06-08-2018, 09:19 PM
Calabrians, Sicilians are primarily Barchin Neolithic or Anatolian EBA with Italic and secondary input from other areas. ie. Aberesche in some regions or Arabic/Maghrebi in some.
Dodecanese are the most Eastern shifted region of Greece (geography speaking would be part of Anatolia rather Europe) so their gedmatch scores aren't that surprising.. Calabria on the other hand is more outlying for their region expecting them to be more Western shifted than Greek Islanders but a lot are just as Near East shifted as the most outlying Greek Islands.

For example a Dodecanese Greek from Karpathos gets very similar to South Italians but have East European admixture instead of the Western European which captures the Continental European ancestry in Sicilians.

# Population Percent
1 SW_Europe 34.13
2 West_Asia 27.62
3 SW_Asia 20.57
4 NE_Europe 13.58
5 Siberia 1.08
6 South_Asia 1.08
7 NE_Asia 0.9
8 West_Africa 0.69
9 Oceania 0.34

Single Population Sharing:

# Population (source) Distance
1 Sephardic_Jew 4.86
2 Ashkenazy_Jew 7.12
3 Italian_Sicilian 7.42
4 Cypriot 9.07
5 Turkish 10.56
6 Greek_Central 11.19
7 Italian_Abruzzo 11.91
8 Syrian 12.09
9 Lebanese_Christian 12.28
10 Turkish_Kayseri 12.34
11 Lebanese_Druze 12.83
12 Lebanese_Muslim 13.61
13 Jordanian 13.76
14 Turkish_Aydin 14.2
15 Palestinian 16.4
16 Greek_Thessaly 16.65
17 Albanian 17.26
18 Azerbaijan_Azeri 18.1
19 Assyrian 18.34
20 Samaritan_Jew 18.35

Mixed Mode Population Sharing:

# Primary Population (source) Secondary Population (source) Distance
1 69.9% Sephardic_Jew + 30.1% Turkish @ 1.91
2 69.4% Lebanese_Muslim + 30.6% Spaniard @ 1.96
3 73.3% Sephardic_Jew + 26.7% Turkish_Kayseri @ 1.96
4 76.1% Sephardic_Jew + 23.9% Turkish_Aydin @ 2
5 51.6% Albanian + 48.4% Samaritan_Jew @ 2.06
6 70.7% Lebanese_Druze + 29.3% Spaniard @ 2.15
7 52.5% Greek_Thessaly + 47.5% Samaritan_Jew @ 2.16
8 71.7% Lebanese_Christian + 28.3% Spaniard @ 2.2
9 66.8% Ashkenazy_Jew + 33.2% Jordanian @ 2.25
10 70.6% Ashkenazy_Jew + 29.4% Palestinian @ 2.26
11 80.7% Sephardic_Jew + 19.3% Azerbaijan_Azeri @ 2.29
12 84.7% Sephardic_Jew + 15.3% Iranian @ 2.3
13 62.5% Greek_Central + 37.5% Samaritan_Jew @ 2.36
14 74.3% Lebanese_Christian + 25.7% French @ 2.38
15 67.6% Lebanese_Christian + 32.4% Italian_Bergamo @ 2.39
16 61.2% Lebanese_Christian + 38.8% Italian_Tuscan @ 2.4
17 84.5% Sephardic_Jew + 15.5% Dagestan_Azeri @ 2.46
18 50.1% Samaritan_Jew + 49.9% Kosovar @ 2.48
19 83.8% Sephardic_Jew + 16.2% Kurdish @ 2.49
20 58.7% Lebanese_Christian + 41.3% Albanian @ 2.5


South Aegean didn't receive any noticeable Jewish or Levantine input still look at how East Mediterranean shifted and this is likely to be Anatolia EBA rather direct Levant ancestry. This sample is basically zero Mainland Greek like and would fall in the Anatolian Greek cluster as expected for East Aegean heavily colonized by Ionian Greeks and lack the Steppe, Balkanic input.

It actually must be Levantine , and not Anatolia EBA, for several reasons. First, Lazardis et al. shows exactly this - that there is Near Eastern admixture among Greeks, and it's more substantial among Aegean Greeks than among mainland Greeks.

Second, look at the distances of what you've posted:


# Population (source) Distance
1 Sephardic_Jew 4.86
2 Ashkenazy_Jew 7.12
3 Italian_Sicilian 7.42
4 Cypriot 9.07
5 Turkish 10.56
6 Greek_Central 11.19
7 Italian_Abruzzo 11.91
8 Syrian 12.09
9 Lebanese_Christian 12.28
10 Turkish_Kayseri 12.34
11 Lebanese_Druze 12.83
12 Lebanese_Muslim 13.61
13 Jordanian 13.76
14 Turkish_Aydin 14.2
15 Palestinian 16.4
16 Greek_Thessaly 16.65
17 Albanian 17.26
18 Azerbaijan_Azeri 18.1
19 Assyrian 18.34
20 Samaritan_Jew 18.35

Western Jews, Christian Lebanese, Syrians and Druze - all Levantines - are pretty high on the list (relatively speaking). Western Jews score HIGHER than other known Aegean-mixed people - such as Sicilians and Cypriots.

According to the model I've found - all of these Levantines have Aegean admixture (Mycenean-like and Minoan-like), of about 30-25%. Western Jews seem to have 50% Aegean admixture. The distances of the Levantine populations you've got actually seem to go according to the amount of Aegean admixture. And where are Samaritans? they are the lowest - at a distance of @ 18. Yet, Samaritans and Jews and Lebanese all derive from the same BA Levantines - the only difference is that Samaritans seem to have zero Aegean admixture (probably because they didn't proselytize and also developed an identity before Christianity - so they didn't mixed with Byzantine Greeks later on). Samaritans even get bigger distance than Assyrians, which are by no means Levantines.

Sikeliot
06-08-2018, 10:39 PM
Calabrians, Sicilians are primarily Barchin Neolithic or Anatolian EBA with Italic and secondary input from other areas. ie. Aberesche in some regions or Arabic/Maghrebi in some.
Dodecanese are the most Eastern shifted region of Greece (geography speaking would be part of Anatolia rather Europe) so their gedmatch scores aren't that surprising.. Calabria on the other hand is more outlying for their region expecting them to be more Western shifted than Greek Islanders but a lot are just as Near East shifted as the most outlying Greek Islands.

For example a Dodecanese Greek from Karpathos gets very similar to South Italians but have East European admixture instead of the Western European which captures the Continental European ancestry in Sicilians.

# Population Percent
1 SW_Europe 34.13
2 West_Asia 27.62
3 SW_Asia 20.57
4 NE_Europe 13.58
5 Siberia 1.08
6 South_Asia 1.08
7 NE_Asia 0.9
8 West_Africa 0.69
9 Oceania 0.34

Single Population Sharing:

# Population (source) Distance
1 Sephardic_Jew 4.86
2 Ashkenazy_Jew 7.12
3 Italian_Sicilian 7.42
4 Cypriot 9.07
5 Turkish 10.56
6 Greek_Central 11.19
7 Italian_Abruzzo 11.91
8 Syrian 12.09
9 Lebanese_Christian 12.28
10 Turkish_Kayseri 12.34
11 Lebanese_Druze 12.83
12 Lebanese_Muslim 13.61
13 Jordanian 13.76
14 Turkish_Aydin 14.2
15 Palestinian 16.4
16 Greek_Thessaly 16.65
17 Albanian 17.26
18 Azerbaijan_Azeri 18.1
19 Assyrian 18.34
20 Samaritan_Jew 18.35

Mixed Mode Population Sharing:

# Primary Population (source) Secondary Population (source) Distance
1 69.9% Sephardic_Jew + 30.1% Turkish @ 1.91
2 69.4% Lebanese_Muslim + 30.6% Spaniard @ 1.96
3 73.3% Sephardic_Jew + 26.7% Turkish_Kayseri @ 1.96
4 76.1% Sephardic_Jew + 23.9% Turkish_Aydin @ 2
5 51.6% Albanian + 48.4% Samaritan_Jew @ 2.06
6 70.7% Lebanese_Druze + 29.3% Spaniard @ 2.15
7 52.5% Greek_Thessaly + 47.5% Samaritan_Jew @ 2.16
8 71.7% Lebanese_Christian + 28.3% Spaniard @ 2.2
9 66.8% Ashkenazy_Jew + 33.2% Jordanian @ 2.25
10 70.6% Ashkenazy_Jew + 29.4% Palestinian @ 2.26
11 80.7% Sephardic_Jew + 19.3% Azerbaijan_Azeri @ 2.29
12 84.7% Sephardic_Jew + 15.3% Iranian @ 2.3
13 62.5% Greek_Central + 37.5% Samaritan_Jew @ 2.36
14 74.3% Lebanese_Christian + 25.7% French @ 2.38
15 67.6% Lebanese_Christian + 32.4% Italian_Bergamo @ 2.39
16 61.2% Lebanese_Christian + 38.8% Italian_Tuscan @ 2.4
17 84.5% Sephardic_Jew + 15.5% Dagestan_Azeri @ 2.46
18 50.1% Samaritan_Jew + 49.9% Kosovar @ 2.48
19 83.8% Sephardic_Jew + 16.2% Kurdish @ 2.49
20 58.7% Lebanese_Christian + 41.3% Albanian @ 2.5


South Aegean didn't receive any noticeable Jewish or Levantine input still look at how East Mediterranean shifted and this is likely to be Anatolia EBA rather direct Levant ancestry. This sample is basically zero Mainland Greek like and would fall in the Anatolian Greek cluster as expected for East Aegean heavily colonized by Ionian Greeks and lack the Steppe, Balkanic input.


What I have noticed is that the North Aegean islands, Cyclades islands, and western Crete are genetically somewhat shifted toward mainland Greece and are similar to Apulians and to the least Near Eastern of Sicilians, while the South Aegean and eastern Crete are closer to Calabria and to the more Near Eastern shifted of Sicilians. But I would say overall both North and South Aegean are still closer to south Italy than to mainland Greece.



It actually must be Levantine , and not Anatolia EBA, for several reasons. First, Lazardis et al. shows exactly this - that there is Near Eastern admixture among Greeks, and it's more substantial among Aegean Greeks than among mainland Greeks.

Second, look at the distances of what you've posted:


# Population (source) Distance
1 Sephardic_Jew 4.86
2 Ashkenazy_Jew 7.12
3 Italian_Sicilian 7.42
4 Cypriot 9.07
5 Turkish 10.56
6 Greek_Central 11.19
7 Italian_Abruzzo 11.91
8 Syrian 12.09
9 Lebanese_Christian 12.28
10 Turkish_Kayseri 12.34
11 Lebanese_Druze 12.83
12 Lebanese_Muslim 13.61
13 Jordanian 13.76
14 Turkish_Aydin 14.2
15 Palestinian 16.4
16 Greek_Thessaly 16.65
17 Albanian 17.26
18 Azerbaijan_Azeri 18.1
19 Assyrian 18.34
20 Samaritan_Jew 18.35

Western Jews, Christian Lebanese, Syrians and Druze - all Levantines - are pretty high on the list (relatively speaking). Western Jews score HIGHER than other known Aegean-mixed people - such as Sicilians and Cypriots.

According to the model I've found - all of these Levantines have Aegean admixture (Mycenean-like and Minoan-like), of about 30-25%. Western Jews seem to have 50% Aegean admixture. The distances of the Levantine populations you've got actually seem to go according to the amount of Aegean admixture. And where are Samaritans? they are the lowest - at a distance of @ 18. Yet, Samaritans and Jews and Lebanese all derive from the same BA Levantines - the only difference is that Samaritans seem to have zero Aegean admixture (probably because they didn't proselytize and also developed an identity before Christianity - so they didn't mixed with Byzantine Greeks later on). Samaritans even get bigger distance than Assyrians, which are by no means Levantines.



For comparison here is a Calabrese person who is on my relatives list:

# Population Percent
1 SW_Europe 36.84
2 West_Asia 24.35
3 SW_Asia 21.2
4 NE_Europe 14.04
5 Oceania 1.08
6 SE_Asia 0.88
7 West_Africa 0.83
8 NE_Asia 0.4
9 East_Africa 0.33
10 Siberia 0.06

Single Population Sharing:

# Population (source) Distance
1 Sephardic_Jew 2.64
2 Italian_Sicilian 5.79
3 Ashkenazy_Jew 5.95
4 Italian_Abruzzo 10.14
5 Greek_Central 10.76
6 Cypriot 11.08
7 Turkish 14.64
8 Lebanese_Christian 15.02
9 Syrian 15.16
10 Greek_Thessaly 15.57
11 Jordanian 15.8
12 Lebanese_Druze 15.97
13 Albanian 16.08
14 Turkish_Kayseri 16.46
15 Lebanese_Muslim 16.73
16 Kosovar 17.35
17 Italian_Tuscan 17.35
18 Turkish_Aydin 17.92
19 Palestinian 18.35
20 Samaritan_Jew 20.09

Mixed Mode Population Sharing:

# Primary Population (source) Secondary Population (source) Distance
1 97.8% Sephardic_Jew + 2.2% Visayan @ 1.95
2 98.6% Sephardic_Jew + 1.4% Papuan @ 1.97
3 97.8% Sephardic_Jew + 2.2% Malayan @ 1.99
4 98% Sephardic_Jew + 2% Luzon @ 2.02
5 97.9% Sephardic_Jew + 2.1% Filipino @ 2.04
6 98.1% Sephardic_Jew + 1.9% Dusun @ 2.05
7 97.7% Sephardic_Jew + 2.3% Cambodian @ 2.05
8 98.1% Sephardic_Jew + 1.9% Murut @ 2.07
9 97.7% Sephardic_Jew + 2.3% Thai @ 2.09
10 98% Sephardic_Jew + 2% Dai @ 2.09
11 98% Sephardic_Jew + 2% Vietnamese @ 2.11
12 98% Sephardic_Jew + 2% Lahu @ 2.13
13 97.8% Sephardic_Jew + 2.2% Kikuyu @ 2.14
14 98.5% Sephardic_Jew + 1.5% Igorot @ 2.15
15 98.4% Sephardic_Jew + 1.6% Gambian @ 2.17
16 97.9% Sephardic_Jew + 2.1% Burmese @ 2.17
17 98.5% Sephardic_Jew + 1.5% Mandinka @ 2.18
18 98.5% Sephardic_Jew + 1.5% Bantu_SW @ 2.18
19 98.4% Sephardic_Jew + 1.6% Kaba @ 2.18
20 98.3% Sephardic_Jew + 1.7% Bantu_SE @ 2.18



Here is a Sicilian from Palermo... similar to the Calabrese and to the Dodecanese person except for they get Tunisian on their list, also which is likely reflective of Arab era mixture:

# Population Percent
1 SW_Europe 38.49
2 West_Asia 24.78
3 SW_Asia 18.27
4 NE_Europe 12.37
5 South_Asia 1.69
6 West_Africa 1.42
7 East_Africa 1.24
8 Siberia 0.75
9 South_Africa 0.59
10 SE_Asia 0.27
11 NE_Asia 0.14

Single Population Sharing:

# Population (source) Distance
1 Sephardic_Jew 3.77
2 Italian_Sicilian 4.14
3 Ashkenazy_Jew 5.37
4 Italian_Abruzzo 9.71
5 Greek_Central 10.63
6 Cypriot 10.78
7 Turkish 14.32
8 Greek_Thessaly 15.34
9 Syrian 15.56
10 Lebanese_Christian 15.67
11 Albanian 15.9
12 Turkish_Kayseri 16.09
13 Lebanese_Druze 16.31
14 Jordanian 16.95
15 Italian_Tuscan 17.1
16 Lebanese_Muslim 17.14
17 Kosovar 17.3
18 Turkish_Aydin 17.75
19 Palestinian 19.66
20 Tunisian 21.13

Mixed Mode Population Sharing:

# Primary Population (source) Secondary Population (source) Distance
1 83.7% Italian_Sicilian + 16.3% Jordanian @ 2.56
2 85.8% Italian_Sicilian + 14.2% Palestinian @ 2.6
3 86.8% Italian_Sicilian + 13.2% Samaritan_Jew @ 2.61
4 88.6% Italian_Sicilian + 11.4% Egyptian_Copts @ 2.67
5 88.6% Italian_Sicilian + 11.4% Egyptian @ 2.75
6 90.7% Italian_Sicilian + 9.3% Yemeni @ 2.85
7 83.7% Italian_Sicilian + 16.3% Syrian @ 2.87
8 83.9% Italian_Sicilian + 16.1% Lebanese_Christian @ 2.91
9 97.3% Sephardic_Jew + 2.7% Luo @ 2.93
10 85.3% Italian_Sicilian + 14.7% Lebanese_Muslim @ 2.94
11 97.4% Sephardic_Jew + 2.6% Bantu_NE @ 2.95
12 97.3% Sephardic_Jew + 2.7% Mada @ 2.95
13 97.4% Sephardic_Jew + 2.6% Bantu_SE @ 2.95
14 97.4% Sephardic_Jew + 2.6% Kaba @ 2.95
15 92.8% Italian_Sicilian + 7.2% Saudi @ 2.97
16 84.8% Italian_Sicilian + 15.2% Lebanese_Druze @ 2.99
17 97.7% Sephardic_Jew + 2.3% Bantu_SW @ 3
18 96.5% Sephardic_Jew + 3.5% Fulani @ 3
19 97.4% Sephardic_Jew + 2.6% Biaka_Pygmy @ 3.01
20 93.8% Italian_Sicilian + 6.2% Bedouin @ 3.02


And here is a Greek from the North Aegean islands (Chios), as you see they are still closer to south Italians overall but they are definitely less Levantine/Near Eastern than Sicilians, Dodecanese, or Ashkenazim.

# Population Percent
1 SW_Europe 35.38
2 West_Asia 27
3 NE_Europe 19.43
4 SW_Asia 14.92
5 Siberia 1.77
6 East_Africa 0.51
7 South_Africa 0.47
8 Oceania 0.24
9 West_Africa 0.16
10 South_Asia 0.13

Single Population Sharing:

# Population (source) Distance
1 Greek_Central 4.17
2 Ashkenazy_Jew 4.4
3 Italian_Sicilian 5.93
4 Italian_Abruzzo 6.57
5 Sephardic_Jew 8.44
6 Greek_Thessaly 9.71
7 Albanian 10.42
8 Kosovar 11.43
9 Turkish 12.47
10 Italian_Tuscan 12.81
11 Turkish_Aydin 14.45
12 Cypriot 14.83
13 Turkish_Kayseri 15.15
14 Bulgarian 15.8
15 Macedonian 16.4
16 Romanian 17.6
17 Balkar 17.88
18 Syrian 18.57
19 Montenegrin 18.58
20 Italian_Bergamo 19.03

Mixed Mode Population Sharing:

# Primary Population (source) Secondary Population (source) Distance
1 69.5% Italian_Abruzzo + 30.5% Turkish_Aydin @ 1.85
2 66.2% Italian_Abruzzo + 33.8% Turkish @ 1.87
3 80.2% Greek_Central + 19.8% Turkish_Kayseri @ 1.91
4 83.8% Greek_Central + 16.2% Lebanese_Druze @ 1.95
5 65.1% Albanian + 34.9% Lebanese_Druze @ 1.97
6 84.4% Greek_Central + 15.6% Lebanese_Muslim @ 1.97
7 83.4% Greek_Central + 16.6% Syrian @ 1.99
8 77.2% Greek_Central + 22.8% Turkish @ 2.04
9 77.2% Italian_Abruzzo + 22.8% Ossetian @ 2.05
10 83.9% Greek_Central + 16.1% Lebanese_Christian @ 2.06
11 66.8% Greek_Thessaly + 33.2% Lebanese_Druze @ 2.06
12 83.3% Ashkenazy_Jew + 16.7% Kumyk @ 2.08
13 69.3% Albanian + 30.7% Assyrian @ 2.09
14 62.9% Kosovar + 37.1% Lebanese_Druze @ 2.1
15 66.1% Albanian + 33.9% Lebanese_Muslim @ 2.12
16 76.3% Italian_Abruzzo + 23.7% Azerbaijan_Azeri @ 2.13
17 70.7% Italian_Abruzzo + 29.3% Turkish_Kayseri @ 2.15
18 67.8% Greek_Thessaly + 32.2% Lebanese_Muslim @ 2.16
19 86.6% Greek_Central + 13.4% Assyrian @ 2.19
20 64% Kosovar + 36% Lebanese_Muslim @ 2.2



According to the model I've found - all of these Levantines have Aegean admixture (Mycenean-like and Minoan-like), of about 30-25%. Western Jews seem to have 50% Aegean admixture. The distances of the Levantine populations you've got actually seem to go according to the amount of Aegean admixture. And where are Samaritans? they are the lowest - at a distance of @ 18. Yet, Samaritans and Jews and Lebanese all derive from the same BA Levantines - the only difference is that Samaritans seem to have zero Aegean admixture (probably because they didn't proselytize and also developed an identity before Christianity - so they didn't mixed with Byzantine Greeks later on). Samaritans even get bigger distance than Assyrians, which are by no means Levantines.


Again, Aegean people were the recipient of a wave of migration also populating southern Italy and the Levant, from Anatolia. Calling it "Aegean" ancestry incorrectly portrays the direction of the gene flow.

jonahst
06-08-2018, 11:11 PM
I'm also not claiming that there is no Levantine ancestry to Western Jews. On the contrary - I claim it's about 30%-35%.

I think it's important to make a distinction here between Bronze Age Levantine ancestry and Second Temple Levantine ancestry. From what I gather, part of your argument is that some, if not most, of the theoretical Aegean admixture in Western Jews entered while still in the Levant. Therefore, Second Temple Levantine would already include this admixture, and presumably it would have similarly impacted other Levantine groups (except Samaritans), right? So, while it definitely might be true that Western Jews are 30-35% Bronze Age Levantine, if your theory is true, then they are still 50-60% Second Temple Levantine.

The genetic makeup of "Levantine" has always been changing. Obviously as you've shown, Bronze Age Levant is quite different from Neolithic Levant, which is quite different from Natufian.

Maybe this seems obvious, but I just wanted to point it out...

jonahst
06-08-2018, 11:12 PM
Double post

Sikeliot
06-08-2018, 11:12 PM
There's only one branch of J1 among Ashkenazi Jews which can be securely labeled Arabian, namely ZS2102 which is a branch of FGC8712 exactly like L859 (the main Hashemite branch). This might actually be Nabatean in origin, if it isn't older that is.

But it does seem like Jews (Ashkenazim and Sephardim) as well as Sicilians/Calabrese and Cypriots have some level of SW Asian/Arabian ancestry that cannot be explained without either Arabian ancestry, or their Levantine source population being on the higher end for SW Asian. Especially Ashkenazim and Sicilians, whose SW Asian to Caucasus ratio implies a SW Asian-heavy input.

jonahst
06-08-2018, 11:17 PM
no ;)
we have levantine ancestery not arabian
i score 0% arabian in livingdna but i do score significant levantine
same goes for johns who is full aschenazi he score only 1-1.5% arabia in living dna extremely low .....

In all fairness, a Bahraini girl also got 0% Arabian. I wouldn't take their labels too literally :P

Sikeliot
06-08-2018, 11:18 PM
Calabrians, Sicilians are primarily Barchin Neolithic or Anatolian EBA with Italic and secondary input from other areas. ie. Aberesche in some regions or Arabic/Maghrebi in some.

I think this is where we see a difference between northeastern, southeastern, and western-central Sicily. Southeast Sicily clearly has elevated Steppe compared to the rest of the island which could very well have come from Doric Greeks and/or a more recent Greek migration that already contained Slavic ancestry -- the same would be true for Apulia. This makes sense in the context of knowing that under Arab rule, the southeast of Sicily was still heavily Greek speaking and Christian and received continued Greek settlement, and was also the region that in ancient times had never been Phoenician nor Carthaginian.

It is also clear that, even if you maintain they have absorbed some admixture, northeast Sicily has a more Neolithic genetic profile more consistent with an older population strain that was heavy in Caucasus and Anatolian Neolithic ancestry while not having as much SW Asian/North African/Arabian. If you see genetic results from western and central Sicily, there are some people who have quite high SW Asian/North African, but lower Caucasus, which would imply ancestry from a population high in Afroasiatic ancestry but low in Caucasus. This could be North Africans, Arabians, southern Levantines, or all of the above or even a migration we do not know about yet. But it is clear that of the three corners of Sicily, the northeast has absorbed comparatively less foreign input.


In all fairness, a Bahraini girl also got 0% Arabian. I wouldn't take their labels too literally :P

Yes. LivingDNA is not the most reliable of the tests in my view.

Sikeliot
06-08-2018, 11:20 PM
I think it's important to make a distinction here between Bronze Age Levantine ancestry and Second Temple Levantine ancestry. From what I gather, part of your argument is that some, if not most, of the theoretical Aegean admixture in Western Jews entered while still in the Levant. Therefore, Second Temple Levantine would already include this admixture, and presumably it would have similarly impacted other Levantine groups (except Samaritans), right? So, while it definitely might be true that Western Jews are 30-35% Bronze Age Levantine, if your theory is true, then they are still 50-60% Second Temple Levantine.

The genetic makeup of "Levantine" has always been changing. Obviously as you're shown, Bronze Age Levant is quite different from Neolithic Levant, which is quite different from Natufian.

Maybe this seems obvious, but I just wanted to point it out...


There is also a difference between the northern and southern Levant, likely on a gradient where Caucasus-related ancestry increases as you go north and SW Asian ancestry decreases and vice versa. What I am proposing is Jews descend from a Levantine population that was one of the heavier on the SW Asian side.

Agamemnon
06-08-2018, 11:21 PM
But it does seem like Jews (Ashkenazim and Sephardim) as well as Sicilians/Calabrese and Cypriots have some level of SW Asian/Arabian ancestry that cannot be explained without either Arabian ancestry, or their Levantine source population being on the higher end for SW Asian. Especially Ashkenazim and Sicilians, whose SW Asian to Caucasus ratio implies a SW Asian-heavy input.

It can easily be explained by the fact that the Judean source might have been Samaritan-like, therefore closer to Jordan_EBA than present-day Levantines. There's no real argument to be made in favour of additional Arabian ancestry, at best we're dealing with pretty rare lineages.

Sikeliot
06-09-2018, 12:00 AM
1. Phoenician colonization and settlement in Sicily is well documented, as well as Greek. Sicily and South Italy at one point, had specific Greek dialects directly derive from Greece mainland and Aegean Islands (I'm talking 2000 years ago, yes?). Modern qpAdm modelling of Sicilians always include substantial Levantine admixture - about ~10% Levantine, ~10% North African. Around 37% of it is Greek - as in derived from those Greek colonialists.

I agree that there is Levantine, North African, and in all likelihood Arabian ancestry within Sicily, but I question the notion still of significant mainland Greek input.

I am not disputing that south Italy and Sicily absorbed Greek ancestry but I want you to explain the following... these are all quotations from the Sarno study, and to me, they imply that most of the Hellenization was cultural, with the exception of Apulia, Lucania, and Trapani, unless we are to believe that the Greeks who came to Sicily and Calabria were pretty much unrelated to modern mainland Greeks altogether.

URL: https://www.nature.com/articles/s41598-017-01802-4

"FineSTRUCTURE results reconnect virtually all the individuals from Albanian and Kosovo, as well as the major part of individuals from mainland Greek populations, to a Southern-Balkan specific cluster (cyan in Fig. 3), which is almost completely absent in Greek-speaking islands and Southern Italy (except for Calabrian Arbereshe), instead showing relatively more similarity with Northern Italian populations "

"Interestingly, despite showing much lower values of sharing, some Balkan IBD-relatedness also emerges in Greek-speaking islands as well as in Apulia and Western Sicily, presumptively reproducing some forms of interaction with Greece and the Balkans in the very recent ancestry of these areas, as consistently signalled by a common sharing of individuals in the FineSTRUCTURE AW-Sicily cluster (see also Supplementary Information)."


"While Albanian-speaking Arbereshe trace their recent genetic ancestry to the Southern Balkans, the Greek-speaking communities of both Apulia (Griko) and Calabria (Grecani) show no clear signs of a recent (i.e. from the late Middle Ages) continental Greek origin, instead resembling the ‘continuum’ populations of Southern Italy and the Greek-speaking islands "

"Furthermore, we observed that both Calabrian and Apulian Greeks from Southern Italy almost completely lack the ‘Southern Balkan’ genetic component detected in Continental Greece and Albania, as well as in the Arbereshe. In both cases, this is consistent with the fact that their arrival in Southern Italy should at least predate those population processes associated to the more recent (i.e. late medieval) differentiation of continental Greek and Southern Balkan groups (cf. paragraph below). This does not exclude migrations from Aegean/Dodecanese and Crete islands, that presumptively did not (or only marginally) experienced - by virtue of their higher geographic marginality - the North-South Balkan gene flow that instead interested the continental part of Greece."

"Similar contacts may be also responsible of the relatively high Balkan-IBD sharing (Fig. 4, Supplementary Table S5, Supplementary Information) revealed by Apulia (including Salentino Greeks), West Sicily, and Sicilian Arbereshe (i.e. populations with higher frequencies of the fineSTRUCTURE AW-Sicily cluster) compared to Calabria and Eastern Sicily (CE-Sicily cluster)."

giorgio
06-09-2018, 12:06 AM
It actually must be Levantine , and not Anatolia EBA, for several reasons. First, Lazardis et al. shows exactly this - that there is Near Eastern admixture among Greeks, and it's more substantial among Aegean Greeks than among mainland Greeks.

Second, look at the distances of what you've posted:


# Population (source) Distance
1 Sephardic_Jew 4.86
2 Ashkenazy_Jew 7.12
3 Italian_Sicilian 7.42
4 Cypriot 9.07
5 Turkish 10.56
6 Greek_Central 11.19
7 Italian_Abruzzo 11.91
8 Syrian 12.09
9 Lebanese_Christian 12.28
10 Turkish_Kayseri 12.34
11 Lebanese_Druze 12.83
12 Lebanese_Muslim 13.61
13 Jordanian 13.76
14 Turkish_Aydin 14.2
15 Palestinian 16.4
16 Greek_Thessaly 16.65
17 Albanian 17.26
18 Azerbaijan_Azeri 18.1
19 Assyrian 18.34
20 Samaritan_Jew 18.35

Western Jews, Christian Lebanese, Syrians and Druze - all Levantines - are pretty high on the list (relatively speaking). Western Jews score HIGHER than other known Aegean-mixed people - such as Sicilians and Cypriots.

According to the model I've found - all of these Levantines have Aegean admixture (Mycenean-like and Minoan-like), of about 30-25%. Western Jews seem to have 50% Aegean admixture. The distances of the Levantine populations you've got actually seem to go according to the amount of Aegean admixture. And where are Samaritans? they are the lowest - at a distance of @ 18. Yet, Samaritans and Jews and Lebanese all derive from the same BA Levantines - the only difference is that Samaritans seem to have zero Aegean admixture (probably because they didn't proselytize and also developed an identity before Christianity - so they didn't mixed with Byzantine Greeks later on). Samaritans even get bigger distance than Assyrians, which are by no means Levantines.

This individual is actually more "Levant" shifted than the Ashkenazi average, despite there's no documented Levant input in the Aegean. I guess that there's no clear division between Anatolian-Neolithic Anatolian and Levantine/Arabian ancestry here after all.


Kit M179262

# Population Percent
1 Caucasian 36.16
2 Mediterranean 21.67
3 SW-Asian 17.74
4 NE-Euro 11.53
5 Baloch 10.93
6 S-Indian 1.04
7 NE-Asian 0.35
8 E-African 0.26
9 Siberian 0.25
10 Papuan 0.08

Single Population Sharing:

# Population (source) Distance
1 sephardic-jew (behar) 5.91
2 ashkenazy-jew (behar) 6.7
3 morocco-jew (behar) 8.6
4 ashkenazi (harappa) 9.99
5 cypriot (behar) 13.18
6 turk-aydin (hodoglugil) 13.49
7 lebanese (behar) 13.73
8 turk-kayseri (hodoglugil) 15.84
9 syrian (behar) 16.06
10 palestinian (harappa) 16.21
11 turk-istanbul (hodoglugil) 16.52
12 lebanese-muslim (haber) 16.58
13 turkish (harappa) 17.22
14 jordanian (behar) 17.39
15 turk (behar) 17.52
16 armenian (harappa) 17.6
17 iraqi-arab (harappa) 17.83
18 lebanese-christian (haber) 18.97
19 palestinian (hgdp) 19.22
20 tuscan (1000genomes) 19.25

Mixed Mode Population Sharing:

# Primary Population (source) Secondary Population (source) Distance
1 55.9% palestinian (harappa) + 44.1% tuscan (hapmap) @ 2.51
2 79.5% morocco-jew (behar) + 20.5% lezgin (behar) @ 2.51
3 54.4% palestinian (harappa) + 45.6% tuscan (1000genomes) @ 2.52
4 64.8% iraq-jew (behar) + 35.2% spaniard (1000genomes) @ 2.58
5 78.1% ashkenazy-jew (behar) + 21.9% iraqi-mandaean (harappa) @ 2.6
6 86.1% sephardic-jew (behar) + 13.9% urkarah (xing) @ 2.61
7 80% morocco-jew (behar) + 20% urkarah (xing) @ 2.68
8 57% iraq-jew (behar) + 43% italian (hgdp) @ 2.7
9 56.2% syrian (behar) + 43.8% tuscan (hapmap) @ 2.74
10 72.3% ashkenazy-jew (behar) + 27.7% palestinian (harappa) @ 2.74
11 74.2% ashkenazy-jew (behar) + 25.8% iraqi-arab (harappa) @ 2.75
12 77.4% ashkenazy-jew (behar) + 22.6% iraq-jew (behar) @ 2.76
13 79.3% morocco-jew (behar) + 20.7% chechen (yunusbayev) @ 2.76
14 86% sephardic-jew (behar) + 14% lezgin (behar) @ 2.84
15 83.9% sephardic-jew (behar) + 16.1% stalskoe (xing) @ 2.86
16 76.7% morocco-jew (behar) + 23.3% kumyk (yunusbayev) @ 2.87
17 54.6% syrian (behar) + 45.4% tuscan (1000genomes) @ 2.89
18 78.7% ashkenazy-jew (behar) + 21.3% iranian-jew (behar) @ 2.89
19 66.9% lebanese-christian (haber) + 33.1% spaniard (behar) @ 2.96
20 58.7% lebanese (behar) + 41.3% tuscan (1000genomes) @ 3.06

Sikeliot
06-09-2018, 12:29 AM
This individual is actually more "Levant" shifted than the Ashkenazi average, despite there's no documented Levant input in the Aegean. I guess that there's no clear division between Anatolian-Neolithic Anatolian and Levantine/Arabian ancestry here after all.

Given that Calabria is genetically very similar to the Dodecanese and that it has been hypothesized that Calabria plots how it does due to the assimilation of a once large Jewish population, might this also be the case for the Dodecanese?

J Man
06-09-2018, 12:46 AM
Given that Calabria is genetically very similar to the Dodecanese and that it has been hypothesized that Calabria plots how it does due to the assimilation of a once large Jewish population, might this also be the case for the Dodecanese?

His results have been discussed numerous times before but here are my paternal side grandfather's Eurogenes K13 results for this thread for comparison sake. He comes from the Cosenza province in Calabria.

Admix Results (sorted):

# Population Percent
1 East_Med 34.86
2 West_Med 19.35
3 North_Atlantic 17.11
4 West_Asian 16.13
5 Baltic 6.85
6 Red_Sea 4.31


Finished reading population data. 204 populations found.
13 components mode.

--------------------------------

Least-squares method.

Using 1 population approximation:
1 South_Italian @ 4.972449
2 East_Sicilian @ 6.587192
3 Central_Greek @ 7.088356
4 Ashkenazi @ 7.378014
5 Sephardic_Jewish @ 7.997984
6 Italian_Jewish @ 8.485114
7 Italian_Abruzzo @ 9.341055
8 West_Sicilian @ 9.821878
9 Algerian_Jewish @ 9.978148
10 Tunisian_Jewish @ 12.297909
11 Libyan_Jewish @ 13.327330
12 Greek_Thessaly @ 14.341321
13 Cyprian @ 14.799298
14 Tuscan @ 17.051533
15 Lebanese_Muslim @ 18.396666
16 Syrian @ 20.168324
17 Turkish @ 21.088888
18 Lebanese_Druze @ 22.316444
19 Samaritan @ 22.370014
20 Jordanian @ 22.699770

Using 2 populations approximation:
1 50% Lebanese_Druze +50% North_Italian @ 3.479847


Using 3 populations approximation:
1 50% Italian_Abruzzo +25% Lebanese_Druze +25% South_Italian @ 2.558497


Using 4 populations approximation:
++++++++++++++++++++++++++++++++++++++++++++++++++ +++++++++++++++++++++++++++++++++++
1 Italian_Abruzzo + Italian_Abruzzo + Lebanese_Druze + South_Italian @ 2.558497
2 Cyprian + Italian_Abruzzo + Lebanese_Druze + North_Italian @ 2.563801
3 Central_Greek + Lebanese_Druze + South_Italian + Tuscan @ 2.565588
4 Italian_Abruzzo + Lebanese_Druze + South_Italian + West_Sicilian @ 2.580211
5 Central_Greek + Italian_Abruzzo + Lebanese_Druze + West_Sicilian @ 2.639919
6 Lebanese_Druze + South_Italian + South_Italian + Tuscan @ 2.710793
7 Italian_Abruzzo + Lebanese_Druze + South_Italian + Tuscan @ 2.842303
8 Central_Greek + Cyprian + Lebanese_Druze + Spanish_Valencia @ 2.878429
9 Central_Greek + Cyprian + Lebanese_Druze + Spanish_Cataluna @ 2.890379
10 East_Sicilian + Lebanese_Druze + South_Italian + Tuscan @ 2.901598
11 Italian_Abruzzo + Italian_Abruzzo + Lebanese_Druze + West_Sicilian @ 2.917646
12 East_Sicilian + Italian_Abruzzo + Lebanese_Druze + West_Sicilian @ 2.922808
13 Central_Greek + Italian_Abruzzo + Lebanese_Druze + South_Italian @ 2.924089
14 Central_Greek + Italian_Abruzzo + Italian_Abruzzo + Lebanese_Druze @ 2.927878
15 East_Sicilian + Italian_Abruzzo + Italian_Abruzzo + Lebanese_Druze @ 2.969350
16 Cyprian + Lebanese_Druze + South_Italian + Spanish_Cataluna @ 2.971546
17 Cyprian + Lebanese_Druze + Tuscan + Tuscan @ 2.972785
18 Central_Greek + Lebanese_Druze + West_Sicilian + West_Sicilian @ 2.987027
19 Cyprian + Italian_Abruzzo + Italian_Abruzzo + Italian_Jewish @ 3.015225
20 Central_Greek + Lebanese_Druze + Lebanese_Druze + Southwest_French @ 3.031736

Done.

Elapsed time 0.7004 seconds.

Tz85
06-09-2018, 01:05 AM
His results have been discussed numerous times before but here are my paternal side grandfather's Eurogenes K13 results for this thread for comparison sake. He comes from the Cosenza province in Calabria.

Admix Results (sorted):

# Population Percent
1 East_Med 34.86
2 West_Med 19.35
3 North_Atlantic 17.11
4 West_Asian 16.13
5 Baltic 6.85
6 Red_Sea 4.31


Finished reading population data. 204 populations found.
13 components mode.

--------------------------------

Least-squares method.

Using 1 population approximation:
1 South_Italian @ 4.972449
2 East_Sicilian @ 6.587192
3 Central_Greek @ 7.088356
4 Ashkenazi @ 7.378014
5 Sephardic_Jewish @ 7.997984
6 Italian_Jewish @ 8.485114
7 Italian_Abruzzo @ 9.341055
8 West_Sicilian @ 9.821878
9 Algerian_Jewish @ 9.978148
10 Tunisian_Jewish @ 12.297909
11 Libyan_Jewish @ 13.327330
12 Greek_Thessaly @ 14.341321
13 Cyprian @ 14.799298
14 Tuscan @ 17.051533
15 Lebanese_Muslim @ 18.396666
16 Syrian @ 20.168324
17 Turkish @ 21.088888
18 Lebanese_Druze @ 22.316444
19 Samaritan @ 22.370014
20 Jordanian @ 22.699770

Using 2 populations approximation:
1 50% Lebanese_Druze +50% North_Italian @ 3.479847


Using 3 populations approximation:
1 50% Italian_Abruzzo +25% Lebanese_Druze +25% South_Italian @ 2.558497


Using 4 populations approximation:
++++++++++++++++++++++++++++++++++++++++++++++++++ +++++++++++++++++++++++++++++++++++
1 Italian_Abruzzo + Italian_Abruzzo + Lebanese_Druze + South_Italian @ 2.558497
2 Cyprian + Italian_Abruzzo + Lebanese_Druze + North_Italian @ 2.563801
3 Central_Greek + Lebanese_Druze + South_Italian + Tuscan @ 2.565588
4 Italian_Abruzzo + Lebanese_Druze + South_Italian + West_Sicilian @ 2.580211
5 Central_Greek + Italian_Abruzzo + Lebanese_Druze + West_Sicilian @ 2.639919
6 Lebanese_Druze + South_Italian + South_Italian + Tuscan @ 2.710793
7 Italian_Abruzzo + Lebanese_Druze + South_Italian + Tuscan @ 2.842303
8 Central_Greek + Cyprian + Lebanese_Druze + Spanish_Valencia @ 2.878429
9 Central_Greek + Cyprian + Lebanese_Druze + Spanish_Cataluna @ 2.890379
10 East_Sicilian + Lebanese_Druze + South_Italian + Tuscan @ 2.901598
11 Italian_Abruzzo + Italian_Abruzzo + Lebanese_Druze + West_Sicilian @ 2.917646
12 East_Sicilian + Italian_Abruzzo + Lebanese_Druze + West_Sicilian @ 2.922808
13 Central_Greek + Italian_Abruzzo + Lebanese_Druze + South_Italian @ 2.924089
14 Central_Greek + Italian_Abruzzo + Italian_Abruzzo + Lebanese_Druze @ 2.927878
15 East_Sicilian + Italian_Abruzzo + Italian_Abruzzo + Lebanese_Druze @ 2.969350
16 Cyprian + Lebanese_Druze + South_Italian + Spanish_Cataluna @ 2.971546
17 Cyprian + Lebanese_Druze + Tuscan + Tuscan @ 2.972785
18 Central_Greek + Lebanese_Druze + West_Sicilian + West_Sicilian @ 2.987027
19 Cyprian + Italian_Abruzzo + Italian_Abruzzo + Italian_Jewish @ 3.015225
20 Central_Greek + Lebanese_Druze + Lebanese_Druze + Southwest_French @ 3.031736

Done.

Elapsed time 0.7004 seconds.

Very interesting results. Here are the results of my mom. 50% Southern Italian Marano Marchesato, Cosenza, Calabria (Sephardic Settlement), and Pratola Peligna, Abruzzo, and 50% Polish roughly. You can clearly see the Jewish admixture coming from Calabria.

Admix Results (sorted):

# Population Percent
1 North_Atlantic 20.36
2 East_Med 20.04
3 Baltic 19.59
4 West_Med 16.09
5 West_Asian 9.98
6 East_Asian 4.10
7 Northeast_African 3.19
8 South_Asian 2.05
9 Sub-Saharan 1.26
10 Red_Sea 1.19


Finished reading population data. 204 populations found.
13 components mode.

--------------------------------

Least-squares method.

Using 1 population approximation:
1 Bulgarian @ 7.546438
2 Romanian @ 8.707111
3 Greek_Thessaly @ 9.982306
4 Serbian @ 12.409847
5 Tuscan @ 16.146942
6 Italian_Abruzzo @ 16.290070
7 Central_Greek @ 16.912546
8 West_Sicilian @ 17.290131
9 East_Sicilian @ 17.722164
10 North_Italian @ 17.922604
11 Moldavian @ 18.155870
12 Ashkenazi @ 19.297771
13 Hungarian @ 20.848242
14 South_Italian @ 21.245537
15 Croatian @ 21.531506
16 Austrian @ 23.861099
17 Portuguese @ 24.534725
18 Spanish_Galicia @ 25.433926
19 Spanish_Extremadura @ 25.519445
20 French @ 25.700773

Using 2 populations approximation:
1 50% Ashkenazi +50% Moldavian @ 6.243006


Using 3 populations approximation:
1 50% East_Sicilian +25% Greek_Thessaly +25% La_Brana-1 @ 5.156192


Using 4 populations approximation:
++++++++++++++++++++++++++++++++++++++++++++++++++ ++++++++++++++++++++++++++++++++++++++++++++++++++ +++++++++++++
1 Algerian_Jewish + Bulgarian + East_Sicilian + La_Brana-1 @ 5.150026
2 East_Sicilian + East_Sicilian + Greek_Thessaly + La_Brana-1 @ 5.156192
3 Algerian_Jewish + Bulgarian + Central_Greek + La_Brana-1 @ 5.163017
4 Algerian_Jewish + Bulgarian + La_Brana-1 + South_Italian @ 5.187834
5 Central_Greek + East_Sicilian + Greek_Thessaly + La_Brana-1 @ 5.229041
6 Algerian_Jewish + Greek_Thessaly + Greek_Thessaly + La_Brana-1 @ 5.252636
7 Bulgarian + La_Brana-1 + South_Italian + South_Italian @ 5.281569
8 Algerian_Jewish + La_Brana-1 + Romanian + South_Italian @ 5.297127
9 Greek_Thessaly + Greek_Thessaly + La_Brana-1 + South_Italian @ 5.302997
10 Bulgarian + East_Sicilian + La_Brana-1 + South_Italian @ 5.311954
11 East_Sicilian + Greek_Thessaly + La_Brana-1 + South_Italian @ 5.313483
12 Algerian_Jewish + East_Sicilian + La_Brana-1 + Romanian @ 5.332044
13 Algerian_Jewish + Central_Greek + La_Brana-1 + Romanian @ 5.362298
14 Central_Greek + Greek_Thessaly + La_Brana-1 + South_Italian @ 5.386769
15 Central_Greek + Central_Greek + Greek_Thessaly + La_Brana-1 @ 5.410214
16 Iranian_Jewish + La_Brana-1 + Moldavian + Sardinian @ 5.488004
17 Bulgarian + Central_Greek + La_Brana-1 + Libyan_Jewish @ 5.491062
18 East_Sicilian + Greek_Thessaly + Greek_Thessaly + La_Brana-1 @ 5.491326
19 La_Brana-1 + Libyan_Jewish + Moldavian + South_Italian @ 5.506254
20 Algerian_Jewish + La_Brana-1 + Moldavian + South_Italian @ 5.508245

Done.

Sikeliot
06-09-2018, 01:20 AM
Not to repost this again but if you look, you can see that Reggio Calabria, Calabrese Greeks, Catania, and Palermo all have similar levels of Caucasus, Near Eastern, and European, with the latter being low. These should be the regions plotting closest to Jews. Agrigento has elevated Near Eastern compared to the others but breaks from the pattern because they have less Caucasus and a tiny but more of European.

What is interesting here is we can see what makes the Dodecanese outlying is not Near Eastern admixture, but abnormally high Caucasus.

It is too bad Jews are not on this chart for comparison. I will look in the study to see if I can find them on an admixture chart and will post it if so.

https://i.imgur.com/AnYFa5y.png

J Man
06-09-2018, 02:11 AM
Very interesting results. Here are the results of my mom. 50% Southern Italian Marano Marchesato, Cosenza, Calabria (Sephardic Settlement), and Pratola Peligna, Abruzzo, and 50% Polish roughly. You can clearly see the Jewish admixture coming from Calabria.

Admix Results (sorted):

# Population Percent
1 North_Atlantic 20.36
2 East_Med 20.04
3 Baltic 19.59
4 West_Med 16.09
5 West_Asian 9.98
6 East_Asian 4.10
7 Northeast_African 3.19
8 South_Asian 2.05
9 Sub-Saharan 1.26
10 Red_Sea 1.19


Finished reading population data. 204 populations found.
13 components mode.

--------------------------------

Least-squares method.

Using 1 population approximation:
1 Bulgarian @ 7.546438
2 Romanian @ 8.707111
3 Greek_Thessaly @ 9.982306
4 Serbian @ 12.409847
5 Tuscan @ 16.146942
6 Italian_Abruzzo @ 16.290070
7 Central_Greek @ 16.912546
8 West_Sicilian @ 17.290131
9 East_Sicilian @ 17.722164
10 North_Italian @ 17.922604
11 Moldavian @ 18.155870
12 Ashkenazi @ 19.297771
13 Hungarian @ 20.848242
14 South_Italian @ 21.245537
15 Croatian @ 21.531506
16 Austrian @ 23.861099
17 Portuguese @ 24.534725
18 Spanish_Galicia @ 25.433926
19 Spanish_Extremadura @ 25.519445
20 French @ 25.700773

Using 2 populations approximation:
1 50% Ashkenazi +50% Moldavian @ 6.243006


Using 3 populations approximation:
1 50% East_Sicilian +25% Greek_Thessaly +25% La_Brana-1 @ 5.156192


Using 4 populations approximation:
++++++++++++++++++++++++++++++++++++++++++++++++++ ++++++++++++++++++++++++++++++++++++++++++++++++++ +++++++++++++
1 Algerian_Jewish + Bulgarian + East_Sicilian + La_Brana-1 @ 5.150026
2 East_Sicilian + East_Sicilian + Greek_Thessaly + La_Brana-1 @ 5.156192
3 Algerian_Jewish + Bulgarian + Central_Greek + La_Brana-1 @ 5.163017
4 Algerian_Jewish + Bulgarian + La_Brana-1 + South_Italian @ 5.187834
5 Central_Greek + East_Sicilian + Greek_Thessaly + La_Brana-1 @ 5.229041
6 Algerian_Jewish + Greek_Thessaly + Greek_Thessaly + La_Brana-1 @ 5.252636
7 Bulgarian + La_Brana-1 + South_Italian + South_Italian @ 5.281569
8 Algerian_Jewish + La_Brana-1 + Romanian + South_Italian @ 5.297127
9 Greek_Thessaly + Greek_Thessaly + La_Brana-1 + South_Italian @ 5.302997
10 Bulgarian + East_Sicilian + La_Brana-1 + South_Italian @ 5.311954
11 East_Sicilian + Greek_Thessaly + La_Brana-1 + South_Italian @ 5.313483
12 Algerian_Jewish + East_Sicilian + La_Brana-1 + Romanian @ 5.332044
13 Algerian_Jewish + Central_Greek + La_Brana-1 + Romanian @ 5.362298
14 Central_Greek + Greek_Thessaly + La_Brana-1 + South_Italian @ 5.386769
15 Central_Greek + Central_Greek + Greek_Thessaly + La_Brana-1 @ 5.410214
16 Iranian_Jewish + La_Brana-1 + Moldavian + Sardinian @ 5.488004
17 Bulgarian + Central_Greek + La_Brana-1 + Libyan_Jewish @ 5.491062
18 East_Sicilian + Greek_Thessaly + Greek_Thessaly + La_Brana-1 @ 5.491326
19 La_Brana-1 + Libyan_Jewish + Moldavian + South_Italian @ 5.506254
20 Algerian_Jewish + La_Brana-1 + Moldavian + South_Italian @ 5.508245

Done.

My grandfather came from Aprigliano in the Cosenza province.

kingjohn
06-09-2018, 03:27 AM
how can we be part Arabian if our midlle eastern autosomal dna left the Levant before 7th century ?

eolien
06-09-2018, 07:32 AM
how can we be part Arabian if our midlle eastern autosomal dna left the Levant before 7th century ?

contrary to the common belief about arabs in europe, arabs existed much before the 7th century in the levant. you can associate them with populations that were not settled and were living in the fringes of the civilized world, often in the more arid parts of northern arabia such as sinai, jordan and syria. many were christian and some could have been jewish before.

Erikl86
06-09-2018, 07:45 AM
This individual is actually more "Levant" shifted than the Ashkenazi average, despite there's no documented Levant input in the Aegean.

Not true at all - there is tons of documentation and archeological sites strating from Phoenician settlements in the Aegean.

Phoenician main trade routes AND posts around the 8th century BC (I've circled in blue the Aegean areas):
23810

Phoenician settlements in Crete, dated to the 10th century BC:
https://www.academia.edu/7984678/Phoenician_Presence_at_Iron_Age_Kommos

https://www.kommosconservancy.org/a-cretan-conundrum-2/

Phoenician settlements in Ialyssos, Rhodes have also been documented by the classical philosopher Zeno, and Phoenician pottery has been found at some major Rhodian cemeteries, such as the cemeteries of Exochi, Camiros and Ialyssos.

Even genetic evdience supports this:

"A study by Pierre Zalloua and others (2008) found that six subclades of haplogroup J2 (J-M172) J2 in particular, were "a Phoenician signature" amongst modern male populations tested in "the coastal Lebanese Phoenician Heartland and the broader area of the rest of the Levant (the "Phoenician Periphery")", followed by "Cyprus and South Turkey; then Crete; then Malta and East Sicily; then South Sardinia, Ibiza, and Southern Spain; and, finally, Coastal Tunisia and cities like Tingris in Morocco". "
Source - wikipedia.


And, according to Lazirdis study from last year, all Near Eastern admixture in Greeks is dated from the 8th century BC onward, with many also during Byzantine period (probably Christian Levantines escaping clashes and destruction by centuries of wars between Byzantine Empire and Persian Empires).

So the genetic and archaeological evidence is there.


I guess that there's no clear division between Anatolian-Neolithic Anatolian and Levantine/Arabian ancestry here after all.

Of course there is. Before I'll get to that, I'll just add the graph I've provided from Lazirdis et al study:
23811

This graph is constantly being abused and mis-interpreted here (not by you, but others). The Caucasian admixture in those bars, the yellow one, represents Anatolian admixture. The green bar, represents Sardinian-like admixture which stems from EEF (Early European Farmers). The blue one is from later North/Eastern European admixture (also Steppes people), and the red one represents Levantine admixture. These are different, distinguished admixture components. There is clear division between the Anatolian and Levantine admixture - this is obvious.

Also, let's examine again the results for this Dodecanese Greek fellow (thanks for providing the kit, btw):


Kit M179262

Single Population Sharing:

# Population (source) Distance
1 sephardic-jew (behar) 5.91
2 ashkenazy-jew (behar) 6.7
3 morocco-jew (behar) 8.6
4 ashkenazi (harappa) 9.99
5 cypriot (behar) 13.18
6 turk-aydin (hodoglugil) 13.49
7 lebanese (behar) 13.73
8 turk-kayseri (hodoglugil) 15.84
9 syrian (behar) 16.06
10 palestinian (harappa) 16.21
11 turk-istanbul (hodoglugil) 16.52
12 lebanese-muslim (haber) 16.58
13 turkish (harappa) 17.22
14 jordanian (behar) 17.39
15 turk (behar) 17.52
16 armenian (harappa) 17.6
17 iraqi-arab (harappa) 17.83
18 lebanese-christian (haber) 18.97
19 palestinian (hgdp) 19.22
20 tuscan (1000genomes) 19.25


And let's go by your logic, for one minute, that "there's no clear division between Anatolian-Neolithic Anatolian and Levantine/Arabian ancestry".
Now regardless of my theory of substantial Aegean admixture for Western Jews, there is a scientific consensus that modern Lebanese, Syrian etc. are more Levantine than Western Jews, right? So by your logic, that would mean the distance between Lebanese and Syrians and that Dodecanese Greek should have been smaller than the distance between him and Western Jews - after all, even if we go by current scientific consensus,
Amount of Levantine ancestry in Lebanese > amount of Levantine ancestry among Western Jews.

Yet - we see here something else completely - populations with less Levantine ancestry turn out to be closer to this Dodecanese Greek than people with more Levantine ancestry. Which should put another nail in the coffin of the claim that Western Jews' genetic affiliation with Aegeans is the result of "no clear division between Anatolian-Neolithic Anatolian and Levantine/Arabian ancestry", which as I've shown has no scientific merit at all, and open up the possibility (along with my own models and findings, and historical records), that this is due Western Jews having higher Aegean admixture than Lebanese or Syrians.

kingjohn
06-09-2018, 07:53 AM
contrary to the common belief about arabs in europe, arabs existed much before the 7th century in the levant. you can associate them with populations that were not settled and were living in the fringes of the civilized world, often in the more arid parts of northern arabia such as sinai, jordan and syria. many were christian and some could have been jewish before.

coluld be but unlikely .... in my opinion
our middle eastern genes resemble lebanese christians and samritans
not saudi arabians

Erikl86
06-09-2018, 07:55 AM
Very interesting results. Here are the results of my mom. 50% Southern Italian Marano Marchesato, Cosenza, Calabria (Sephardic Settlement), and Pratola Peligna, Abruzzo, and 50% Polish roughly. You can clearly see the Jewish admixture coming from Calabria.

Admix Results (sorted):

# Population Percent
1 North_Atlantic 20.36
2 East_Med 20.04
3 Baltic 19.59
4 West_Med 16.09
5 West_Asian 9.98
6 East_Asian 4.10
7 Northeast_African 3.19
8 South_Asian 2.05
9 Sub-Saharan 1.26
10 Red_Sea 1.19


Finished reading population data. 204 populations found.
13 components mode.

--------------------------------

Least-squares method.

Using 1 population approximation:
1 Bulgarian @ 7.546438
2 Romanian @ 8.707111
3 Greek_Thessaly @ 9.982306
4 Serbian @ 12.409847
5 Tuscan @ 16.146942
6 Italian_Abruzzo @ 16.290070
7 Central_Greek @ 16.912546
8 West_Sicilian @ 17.290131
9 East_Sicilian @ 17.722164
10 North_Italian @ 17.922604
11 Moldavian @ 18.155870
12 Ashkenazi @ 19.297771
13 Hungarian @ 20.848242
14 South_Italian @ 21.245537
15 Croatian @ 21.531506
16 Austrian @ 23.861099
17 Portuguese @ 24.534725
18 Spanish_Galicia @ 25.433926
19 Spanish_Extremadura @ 25.519445
20 French @ 25.700773

Using 2 populations approximation:
1 50% Ashkenazi +50% Moldavian @ 6.243006


Using 3 populations approximation:
1 50% East_Sicilian +25% Greek_Thessaly +25% La_Brana-1 @ 5.156192


Using 4 populations approximation:
++++++++++++++++++++++++++++++++++++++++++++++++++ ++++++++++++++++++++++++++++++++++++++++++++++++++ +++++++++++++
1 Algerian_Jewish + Bulgarian + East_Sicilian + La_Brana-1 @ 5.150026
2 East_Sicilian + East_Sicilian + Greek_Thessaly + La_Brana-1 @ 5.156192
3 Algerian_Jewish + Bulgarian + Central_Greek + La_Brana-1 @ 5.163017
4 Algerian_Jewish + Bulgarian + La_Brana-1 + South_Italian @ 5.187834
5 Central_Greek + East_Sicilian + Greek_Thessaly + La_Brana-1 @ 5.229041
6 Algerian_Jewish + Greek_Thessaly + Greek_Thessaly + La_Brana-1 @ 5.252636
7 Bulgarian + La_Brana-1 + South_Italian + South_Italian @ 5.281569
8 Algerian_Jewish + La_Brana-1 + Romanian + South_Italian @ 5.297127
9 Greek_Thessaly + Greek_Thessaly + La_Brana-1 + South_Italian @ 5.302997
10 Bulgarian + East_Sicilian + La_Brana-1 + South_Italian @ 5.311954
11 East_Sicilian + Greek_Thessaly + La_Brana-1 + South_Italian @ 5.313483
12 Algerian_Jewish + East_Sicilian + La_Brana-1 + Romanian @ 5.332044
13 Algerian_Jewish + Central_Greek + La_Brana-1 + Romanian @ 5.362298
14 Central_Greek + Greek_Thessaly + La_Brana-1 + South_Italian @ 5.386769
15 Central_Greek + Central_Greek + Greek_Thessaly + La_Brana-1 @ 5.410214
16 Iranian_Jewish + La_Brana-1 + Moldavian + Sardinian @ 5.488004
17 Bulgarian + Central_Greek + La_Brana-1 + Libyan_Jewish @ 5.491062
18 East_Sicilian + Greek_Thessaly + Greek_Thessaly + La_Brana-1 @ 5.491326
19 La_Brana-1 + Libyan_Jewish + Moldavian + South_Italian @ 5.506254
20 Algerian_Jewish + La_Brana-1 + Moldavian + South_Italian @ 5.508245

Done.

Very interesting. I can also see a lot of Romanian and Bulgarian "scoring" high up - which supports probably migration from Greek mainland during late Byzantine times (or even post-Byzantine times).

Erikl86
06-09-2018, 08:15 AM
I think it's important to make a distinction here between Bronze Age Levantine ancestry and Second Temple Levantine ancestry. From what I gather, part of your argument is that some, if not most, of the theoretical Aegean admixture in Western Jews entered while still in the Levant. Therefore, Second Temple Levantine would already include this admixture, and presumably it would have similarly impacted other Levantine groups (except Samaritans), right? So, while it definitely might be true that Western Jews are 30-35% Bronze Age Levantine, if your theory is true, then they are still 50-60% Second Temple Levantine.

The genetic makeup of "Levantine" has always been changing. Obviously as you've shown, Bronze Age Levant is quite different from Neolithic Levant, which is quite different from Natufian.

Maybe this seems obvious, but I just wanted to point it out...

Yep, precisely my point. I would assume it won't be too different than modern Christian Palestinian ancient Aegean admixture I've found - about 25%, but this is just assumption - maybe Christian Palestinians absorbed some Byzantine Greeks after Jews have already left the region, so it might have been less than what I've got for modern Christian Palestinians. My claim, as you put it well, is that a lot of this Aegean admixture have happened here in the Levant. It was later re-enforced by absorbing proselytized Greco-Romans, which also had substantial ancient Aegean ancestry back then (more than today).

Erikl86
06-09-2018, 08:25 AM
But it does seem like Jews (Ashkenazim and Sephardim) as well as Sicilians/Calabrese and Cypriots have some level of SW Asian/Arabian ancestry that cannot be explained without either Arabian ancestry, or their Levantine source population being on the higher end for SW Asian. Especially Ashkenazim and Sicilians, whose SW Asian to Caucasus ratio implies a SW Asian-heavy input.

As I've shown here several times - can actually be explained very well by Levantine ancestry.

Aegean-derived admixture levels for Sicilians and Western Jews are very close - in some cases Western Jews are even more admixed with Aegeans than Sicilians, which might explain why they sometimes cluster closer to some Greeks than Sicilians. All the models I've tried - for both ancient Aegean (Mycenaean and Minoan) and modern Cypriots (which seem to be about ~65% Aegean-derived, ~35% Levantine, very similar to Western Jews, but without the Northern European admixture), show that Western Jews (Sephardim and Ashkenazim) get around 60-50% Aegean admixture, and about ~30% Levantine admixture.

Cypriots always get about 80-85% with very small distances - of about 0.0X from Western Jews. And 85% of 65/35 ratio (Cypriot admixture) gives you 50/30 ratio - exactly what I've found when I used qpAdm model for BA Levantines and BA Aegeans for Western Jews.

Idwaajeden
06-09-2018, 08:25 AM
Very interesting results. Here are the results of my mom. 50% Southern Italian Marano Marchesato, Cosenza, Calabria (Sephardic Settlement), and Pratola Peligna, Abruzzo, and 50% Polish roughly. You can clearly see the Jewish admixture coming from Calabria.

Admix Results (sorted):

# Population Percent
1 North_Atlantic 20.36
2 East_Med 20.04
3 Baltic 19.59
4 West_Med 16.09
5 West_Asian 9.98
6 East_Asian 4.10
7 Northeast_African 3.19
8 South_Asian 2.05
9 Sub-Saharan 1.26
10 Red_Sea 1.19


Finished reading population data. 204 populations found.
13 components mode.

--------------------------------

Least-squares method.

Using 1 population approximation:
1 Bulgarian @ 7.546438
2 Romanian @ 8.707111
3 Greek_Thessaly @ 9.982306
4 Serbian @ 12.409847
5 Tuscan @ 16.146942
6 Italian_Abruzzo @ 16.290070
7 Central_Greek @ 16.912546
8 West_Sicilian @ 17.290131
9 East_Sicilian @ 17.722164
10 North_Italian @ 17.922604
11 Moldavian @ 18.155870
12 Ashkenazi @ 19.297771
13 Hungarian @ 20.848242
14 South_Italian @ 21.245537
15 Croatian @ 21.531506
16 Austrian @ 23.861099
17 Portuguese @ 24.534725
18 Spanish_Galicia @ 25.433926
19 Spanish_Extremadura @ 25.519445
20 French @ 25.700773

Using 2 populations approximation:
1 50% Ashkenazi +50% Moldavian @ 6.243006


Using 3 populations approximation:
1 50% East_Sicilian +25% Greek_Thessaly +25% La_Brana-1 @ 5.156192


Using 4 populations approximation:
++++++++++++++++++++++++++++++++++++++++++++++++++ ++++++++++++++++++++++++++++++++++++++++++++++++++ +++++++++++++
1 Algerian_Jewish + Bulgarian + East_Sicilian + La_Brana-1 @ 5.150026
2 East_Sicilian + East_Sicilian + Greek_Thessaly + La_Brana-1 @ 5.156192
3 Algerian_Jewish + Bulgarian + Central_Greek + La_Brana-1 @ 5.163017
4 Algerian_Jewish + Bulgarian + La_Brana-1 + South_Italian @ 5.187834
5 Central_Greek + East_Sicilian + Greek_Thessaly + La_Brana-1 @ 5.229041
6 Algerian_Jewish + Greek_Thessaly + Greek_Thessaly + La_Brana-1 @ 5.252636
7 Bulgarian + La_Brana-1 + South_Italian + South_Italian @ 5.281569
8 Algerian_Jewish + La_Brana-1 + Romanian + South_Italian @ 5.297127
9 Greek_Thessaly + Greek_Thessaly + La_Brana-1 + South_Italian @ 5.302997
10 Bulgarian + East_Sicilian + La_Brana-1 + South_Italian @ 5.311954
11 East_Sicilian + Greek_Thessaly + La_Brana-1 + South_Italian @ 5.313483
12 Algerian_Jewish + East_Sicilian + La_Brana-1 + Romanian @ 5.332044
13 Algerian_Jewish + Central_Greek + La_Brana-1 + Romanian @ 5.362298
14 Central_Greek + Greek_Thessaly + La_Brana-1 + South_Italian @ 5.386769
15 Central_Greek + Central_Greek + Greek_Thessaly + La_Brana-1 @ 5.410214
16 Iranian_Jewish + La_Brana-1 + Moldavian + Sardinian @ 5.488004
17 Bulgarian + Central_Greek + La_Brana-1 + Libyan_Jewish @ 5.491062
18 East_Sicilian + Greek_Thessaly + Greek_Thessaly + La_Brana-1 @ 5.491326
19 La_Brana-1 + Libyan_Jewish + Moldavian + South_Italian @ 5.506254
20 Algerian_Jewish + La_Brana-1 + Moldavian + South_Italian @ 5.508245

Done.

I don’t see 50% Polish, or any Polish. This is not Polish

giorgio
06-09-2018, 09:27 AM
As I've shown here several times - can actually be explained very well by Levantine ancestry.

Aegean-derived admixture levels for Sicilians and Western Jews are very close - in some cases Western Jews are even more admixed with Aegeans than Sicilians, which might explain why they sometimes cluster closer to some Greeks than Sicilians. All the models I've tried - for both ancient Aegean (Mycenaean and Minoan) and modern Cypriots (which seem to be about ~65% Aegean-derived, ~35% Levantine, very similar to Western Jews, but without the Northern European admixture), show that Western Jews (Sephardim and Ashkenazim) get around 60-50% Aegean admixture, and about ~30% Levantine admixture.

Cypriots always get about 80-85% with very small distances - of about 0.0X from Western Jews. And 85% of 65/35 ratio (Cypriot admixture) gives you 50/30 ratio - exactly what I've found when I used qpAdm model for BA Levantines and BA Aegeans for Western Jews.

Going by both Y-DNA and MTDNA Ashkenazi Jews are very different from both Greeks and Sicilians, altough they're a strogly bottlenecked population but Y-dna vise Ashkenazi Jews resemble Samaritians or Christian Levantines before any European population with extremely low European derived Y-dna haplogroups which is very impressive given they've been living in Europe over a thousand of years or so.

Erikl86
06-09-2018, 09:35 AM
Going by both Y-DNA and MTDNA Ashkenazi Jews are very different from both Greeks and Sicilians, altough they're a strogly bottlenecked population but Y-dna vise Ashkenazi Jews resemble Samaritians or Christian Levantines before any European population with extremely low European derived Y-dna haplogroups which is very impressive given they've been living in Europe over a thousand of years or so.

Yep, proportions of Y-dna as well as types of Y-dna haplogroups put Ashkenazi Jews right in between modern Lebanese sects.

This is amazing, but also consistent with historical records that claimed most converts were women - circumcision for grown men was a huge deter, and also in classic Greek world was considered ugly and self-mutilation by many. This is one of the reasons why early christian church hurried up to exempt new converts from circumcising - and one of the things which accelerated it's spread throughout the Roman world.

Erikl86
06-09-2018, 11:02 AM
Btw, here's a PCA map of Mycenaean and Minoan, relative to Neolithic Anatolian, Levant, and BA Anatolian. See how they are close, but not the same - and BA Anatolian cluster completely different than Levantines, closer to BA Aegeans.

23812

I've also ran nMonte again, totally discarding ancient Aegeans and putting just BA Anatolians:

[1] "distance%=0.1794 / distance=0.001794"

Ashkenazi_Jew:average

Levant_N:I1704 43.40
French_East:French24434 33.65
Iran_ChL:I1661 16.90
Anatolia_BA:I2499 6.05

Bogus results - and look how BA Anatolian get only 6%. Rhineland French get way too high - 33%, and Levantine get 60% - completely off. Also, if you look at the PCA map, it makes sense why French rose so high - Neolithic Anatolians, which BA Anatolians surely would have significant ancestry of, cluster together with EN_Europeans, which French still carry some similarity to.

Compare that to using Mycenaean and Minoan:

[1] "distance%=0.0677 / distance=0.000677"

Ashkenazi_Jew:average

Minoan_Lasithi:I0070 42.35
Iran_ChL:I1661 22.05
French_East:French24434 20.60
Minoan_Odigitria:I9131 7.85
Mycenaean:I9041 4.00
Levant_N:I1704 2.40
Iran_ChL:I1665 0.65

Overall Bronze Age Levant - ~25%.
Ancient Aegeans - ~54%
Eastern French (Rhineland French) - 20.6%

I've also tried to replace the ancient Aegean admixture with BA Anatolians for Christian Lebanese, this is what I got:

[1] "distance%=0.2119 / distance=0.002119"

Lebanese_Christian:average

Levant_N:I1704 38.0
Iran_ChL:I1661 24.5
Anatolia_BA:I2495 24.1
Anatolia_BA:I2683 13.3

This is interesting, as it gives Levantine admixture of 62%, and BA Anatolian admixture of 38%. This is very similar in proportions I got with Mycenaean and Minoans, but the distance is x10 times as big:

[1] "distance%=0.0292 / distance=0.000292"

Lebanese_Christian:average

Minoan_Lasithi:I0070 31.55
Iran_ChL:I1661 25.15
Levant_N:I1704 24.50
Iran_ChL:I1665 12.55
Mycenaean:I9041 4.30
Levant_N:I1699 1.65
Iran_ChL:I1662 0.30

Roughly 65% Bronze Age Levantines, 35% ancient Aegean.


I assume why the proportions are similar for Lebanese, but the distance is much greater (but still small - 0.2%), is the following:

BA Anatolians and BA Aegeans cluster close (especially Mycenaean), as the PCA map show, but probably a lot of the admixture Christian Lebanese got was post 1st century, during Byzantine times, when they mixed with Christian Anatolians to a much larger degree than Jews during Hellenistic and early Roman times. These Anatolians were already pretty mixed with Greek colonialists which had a lot of Mycenaean DNA. That would actually give a good explanation.

Sikeliot
06-09-2018, 11:26 AM
This graph is constantly being abused and mis-interpreted here (not by you, but others). The Caucasian admixture in those bars, the yellow one, represents Anatolian admixture. The green bar, represents Sardinian-like admixture which stems from EEF (Early European Farmers). The blue one is from later North/Eastern European admixture (also Steppes people), and the red one represents Levantine admixture. These are different, distinguished admixture components. There is clear division between the Anatolian and Levantine admixture - this is obvious.


If the red is the Levantine admixture then a new conclusion (which does not make sense to me) is drawn which is that Sicilians and Calabrese are almost as Levantine as Cypriots are, since the red bar is nearly the same. Interestingly enough, the Dodecanese have less than all of these, with an amount more similar to that of Apulians, Trapanese, or south Italian Arbereshe.





BA Anatolians and BA Aegeans cluster close (especially Mycenaean), as the PCA map show, but probably a lot of the admixture Christian Lebanese got was post 1st century, during Byzantine times, when they mixed with Christian Anatolians to a much larger degree than Jews during Hellenistic and early Roman times. These Anatolians were already pretty mixed with Greek colonialists which had a lot of Mycenaean DNA. That would actually give a good explanation.


I am unsure why you have an intention to make everyone Greek but you did not address my post which had proof that there is little to no IBD sharing between mainland Greece and southern Italy. Given this, it would be even less likely to find actual Greek ancestry in the Levant. What about the study that demonstrated Christian Lebanese match the ancient Canaanite samples up to 93%? The similarity to Mycenaeans is that Mycenaeans and Levantines share ancestry from an Anatolian source population, not that ancient Aegean people settled en masse in the Levant.

Sikeliot
06-09-2018, 11:27 AM
I don’t see 50% Polish, or any Polish. This is not Polish

You can see it because if you mix south Italian and Polish, this averages out to the Bulgaria/Thessaly/Serbia area which is what this person is scoring in their top 5.

Idwaajeden
06-09-2018, 11:35 AM
You can see it because if you mix south Italian and Polish, this averages out to the Bulgaria/Thessaly/Serbia area which is what this person is scoring in their top 5.

No sorry, there is no Polish there is local groups that occupy in former Polish empire, although nothing indicating "Polish"

Using 3 populations approximation:
1 50% East_Sicilian +25% Greek_Thessaly +25% La_Brana-1 @ 5.156192 <-- Polish? Not here.

Sikeliot
06-09-2018, 11:43 AM
No sorry, there is no Polish there is local groups that occupy in former Polish empire, although nothing indicating "Polish"

Using 3 populations approximation:
1 50% East_Sicilian +25% Greek_Thessaly +25% La_Brana-1 @ 5.156192 <-- Polish? Not here.

La Brana and part of Greek Thessaly are capturing some of the Polish.

Idwaajeden
06-09-2018, 11:50 AM
La Brana and part of Greek Thessaly are capturing some of the Polish.

La Brana and Greek Thessaly is not Polish

Moldavian is not Polish

Jewish groups lived in Poland though.

giorgio
06-09-2018, 11:54 AM
Btw, here's a PCA map of Mycenaean and Minoan, relative to Neolithic Anatolian, Levant, and BA Anatolian. See how they are close, but not the same - and BA Anatolian cluster completely different than Levantines, closer to BA Aegeans.

23812

I've also ran nMonte again, totally discarding ancient Aegeans and putting just BA Anatolians:

[1] "distance%=0.1794 / distance=0.001794"

Ashkenazi_Jew:average

Levant_N:I1704 43.40
French_East:French24434 33.65
Iran_ChL:I1661 16.90
Anatolia_BA:I2499 6.05

Bogus results - and look how BA Anatolian get only 6%. Rhineland French get way too high - 33%, and Levantine get 60% - completely off. Also, if you look at the PCA map, it makes sense why French rose so high - Neolithic Anatolians, which BA Anatolians surely would have significant ancestry of, cluster together with EN_Europeans, which French still carry some similarity to.

Compare that to using Mycenaean and Minoan:

[1] "distance%=0.0677 / distance=0.000677"

Ashkenazi_Jew:average

Minoan_Lasithi:I0070 42.35
Iran_ChL:I1661 22.05
French_East:French24434 20.60
Minoan_Odigitria:I9131 7.85
Mycenaean:I9041 4.00
Levant_N:I1704 2.40
Iran_ChL:I1665 0.65

Overall Bronze Age Levant - ~25%.
Ancient Aegeans - ~54%
Eastern French (Rhineland French) - 20.6%

I've also tried to replace the ancient Aegean admixture with BA Anatolians for Christian Lebanese, this is what I got:

[1] "distance%=0.2119 / distance=0.002119"

Lebanese_Christian:average

Levant_N:I1704 38.0
Iran_ChL:I1661 24.5
Anatolia_BA:I2495 24.1
Anatolia_BA:I2683 13.3

This is interesting, as it gives Levantine admixture of 62%, and BA Anatolian admixture of 38%. This is very similar in proportions I got with Mycenaean and Minoans, but the distance is x10 times as big:

[1] "distance%=0.0292 / distance=0.000292"

Lebanese_Christian:average

Minoan_Lasithi:I0070 31.55
Iran_ChL:I1661 25.15
Levant_N:I1704 24.50
Iran_ChL:I1665 12.55
Mycenaean:I9041 4.30
Levant_N:I1699 1.65
Iran_ChL:I1662 0.30

Roughly 65% Bronze Age Levantines, 35% ancient Aegean.


I assume why the proportions are similar for Lebanese, but the distance is much greater (but still small - 0.2%), is the following:

BA Anatolians and BA Aegeans cluster close (especially Mycenaean), as the PCA map show, but probably a lot of the admixture Christian Lebanese got was post 1st century, during Byzantine times, when they mixed with Christian Anatolians to a much larger degree than Jews during Hellenistic and early Roman times. These Anatolians were already pretty mixed with Greek colonialists which had a lot of Mycenaean DNA. That would actually give a good explanation.

Using North Italians instead of French and Levant BA or Sidon instead of the Levant N would increase the accuracy on the first run, despite the larger distance the first breakdown makes more sense. Altough Ashkenazi Jews may come out as Myceanean like I largely doubt they owe the majority of their ancestry to Myceaneans, rather their admixture makes them Myceanean like. Take the example of a modern Mexican who cluster with Central Asians on a PCA plot despite having no Central Asian ancestry at all but his Iberian and Native American end him plotting with Mixed Old World populations.

Sikeliot
06-09-2018, 11:57 AM
Using North Italians instead of French and Levant BA or Sidon instead of the Levant N would increase the accuracy on the first run, despite the larger distance the first breakdown makes more sense. Altough Ashkenazi Jews may come out as Myceanean like I largely doubt they owe the majority of their ancestry to Myceaneans, rather their admixture makes them Myceanean like. Take the example of a modern Mexican who cluster with Central Asians on a PCA plot despite having no Central Asian ancestry at all but his Iberian and Native American end him plotting with Mixed Old World populations.

The other thing people get wrong is they see that Mycenaean samples are closer on Oracles and PCA plots to Ashkenazim/south Italians and assume these groups are predominantly Mycenaean, when what they should be seeing is that those groups all aggregate out to the same place even if their admixture components differ. Mycenaeans did not have as much Levant/Arabian/North African type input as do southern Italians and Ashkenazim but they plot in the same place.

Ashkenazim are also not more Mycenaean than mainland Greeks today, they just look like it on PCAs because of the large amount of Slavic in Greece pulling Greeks away from Mycenaeans.

Tz85
06-09-2018, 11:59 AM
I don’t see 50% Polish, or any Polish. This is not Polish

You do realize you are clueless right? My great mother is 100% Polish from Krakow. This is called genetic pull. When you have someone from Southern Europe who mixed with Eastern Europe, the results get centralized between the two regins.

Sikeliot
06-09-2018, 12:01 PM
You do realize you are clueless right? My great mother is 100% Polish from Krakow. This is called genetic pull. When you have someone from Southern Europe who mixed with Eastern Europe, the results get centralized between the two regins.

South Italian + Polish in varying amounts will make someone plot near northern Greece, Bulgaria, etc. I am close to northern Greeks and Albanians due to my mixture of Iberian, Sicilian, and Polish also.

giorgio
06-09-2018, 12:04 PM
The other thing people get wrong is they see that Mycenaean samples are closer on Oracles and PCA plots to Ashkenazim/south Italians and assume these groups are predominantly Mycenaean, when what they should be seeing is that those groups all aggregate out to the same place even if their admixture components differ. Mycenaeans did not have as much Levant/Arabian/North African type input as do southern Italians and Ashkenazim but they plot in the same place.

Ashkenazim are also not more Mycenaean than mainland Greeks today, they just look like it on PCAs because of the large amount of Slavic in Greece pulling Greeks away from Mycenaeans.

------------- double post

giorgio
06-09-2018, 12:05 PM
The other thing people get wrong is they see that Mycenaean samples are closer on Oracles and PCA plots to Ashkenazim/south Italians and assume these groups are predominantly Mycenaean, when what they should be seeing is that those groups all aggregate out to the same place even if their admixture components differ. Mycenaeans did not have as much Levant/Arabian/North African type input as do southern Italians and Ashkenazim but they plot in the same place.

Ashkenazim are also not more Mycenaean than mainland Greeks today, they just look like it on PCAs because of the large amount of Slavic in Greece pulling Greeks away from Mycenaeans.

I don't see Arabian or North African input in Ashkenazi Jews. They're basically half Samaritians, Christian Palestinians and ~ 1/4-1/3 Myceanean&Roman and the rest can be anything from host populations.

kingjohn
06-09-2018, 12:07 PM
South Italian + Polish in varying amounts will make someone plot near northern Greece, Bulgaria, etc. I am close to northern Greeks and Albanians due to my mixture of Iberian, Sicilian, and Polish also.

so why you are so obsessed with Sicilians
and not the other parts of your heritage ..... ?

giorgio
06-09-2018, 12:07 PM
You do realize you are clueless right? My great mother is 100% Polish from Krakow. This is called genetic pull. When you have someone from Southern Europe who mixed with Eastern Europe, the results get centralized between the two regins.

From where did your Mother get 4% East Asian because if it's Polish then I guess it would be Tatar from Poland rather ethnic Pole but who knows, maybe some Poles are that East Asian admixed in some parts.

Erikl86
06-09-2018, 12:13 PM
If the red is the Levantine admixture then a new conclusion (which does not make sense to me) is drawn which is that Sicilians and Calabrese are almost as Levantine as Cypriots are, since the red bar is nearly the same. Interestingly enough, the Dodecanese have less than all of these, with an amount more similar to that of Apulians, Trapanese, or south Italian Arbereshe.

Why won't it make sense to you? It makes perfect sense - and would also explain, again, why Western Jews cluster so close to Sicilians and S. Italians, and why they get so short distance when modeled with Cypriots.


I am unsure why you have an intention to make everyone Greek but you did not address my post which had proof that there is little to no IBD sharing between mainland Greece and southern Italy. Given this, it would be even less likely to find actual Greek ancestry in the Levant. What about the study that demonstrated Christian Lebanese match the ancient Canaanite samples up to 93%?

I don't know why you claim that there is little to no IBD sharing between mainland Greece and southern Italy, when there's an article from last proving just the opposite:

https://www.nature.com/articles/ejhg201718


By principal component analysis (PCA) and ADMIXTURE analysis the Peloponneseans are clearly distinguishable from the populations of the Slavic homeland and are very similar to Sicilians and Italians. Using a novel method of quantitative analysis of ADMIXTURE output we find that the Slavic ancestry of Peloponnesean subpopulations ranges from 0.2 to 14.4%.

It's not that they don't cluster with S. Italians and Sicilians - it's that they have from 0.2-14.4% Slavic admixture, which make modern Greek Islanders to cluster closer to S. Italians and Sicilians, than mainland Greeks, but mainland Greeks still cluster close to Italians. The more north you go in Greece, the more Slavic admixture you encounter. Greek Islanders are least admixed with Slavic immigrants, while the vast MAJORITY of Greek ancestry in Italy and Sicily predates the arrival of Slavic tribes to Greece, even if later on few Greek mainlanders also arrived to Italy, they were by far a minority. It's quite obvious.

As for the study, when I just run Hebar et al. own qpAdm model for BA Levantines, against Lebanese Christians, I get the following results:

[1] "distance%=3.8862 / distance=0.038862"

Lebanese_Christian:average

Levant_N:I1704 64.8
Iran_ChL:I1674 35.2

Compare that to the Samaritan results below. Obviously, it's close - we're talking about less than 3.8 here - which would make sense since Lebanese are mostly Levantines according to my own results, but there's something missing, because by adding BA Aegeans, I get much, much closer results.



The similarity to Mycenaeans is that Mycenaeans and Levantines share ancestry from an Anatolian source population, not that ancient Aegean people settled en masse in the Levant.

A good control group would be Samaritans. We all agree Samaritans should be good representative to ancient Canaanites - better than modern-day Lebanese or Western Jews.

This is what I got, when tried to put BA Anatolians instead of BA Aegeans:

[1] "distance%=0.5681 / distance=0.005681"

Samaritan:GSM537032

Iran_ChL:I1665 34.5
Levant_N:I1704 33.2
Levant_N:I0867 32.4
Iran_ChL:I1661 0.0
Iran_ChL:I1662 0.0
Iran_ChL:I1674 0.0
Levant_N:I1699 0.0
Anatolia_BA:I2495 0.0
Anatolia_BA:I2499 0.0
Anatolia_BA:I2683 0.0

Again, zeroed out. This is probably because of isolation and genetic drift, but also due to zero admixture with classic Aegeans - which were close to BA Anatolians.

Let's check BA Levant as reference:

[1] "distance%=0.17657 / distance=0.0017657"

Levant_BA:average

Levant_N:I0867 75.9
Iran_ChL:I1665 24.1
Iran_ChL:I1661 0.0
Iran_ChL:I1662 0.0
Iran_ChL:I1674 0.0
Levant_N:I1699 0.0
Levant_N:I1704 0.0
Anatolia_BA:I2495 0.0
Anatolia_BA:I2499 0.0
Anatolia_BA:I2683 0.0


Seems pretty consistent with Hebar et al. qpAdm model of BA Levantines. It also rules out any significant BA Anatolian ancestry to BA Levantines, despite my assumption earlier that it would exist from Hitties or Hurrians etc..

And now, let's do a mix of both BA Aegeans, and BA Anatolians, and qpAdm of BA Levantines, for Lebanese Christians:

Full nMonte (to show that BA Anatolian alone got 0):

[1] "distance%=0.0339 / distance=0.000339"

Lebanese_Christian:average

Iran_ChL:I1661 29.40
Levant_N:I1704 27.05
Minoan_Lasithi:I0070 25.75
Minoan_Lasithi:I0073 3.30
Iran_ChL:I1665 3.20
Mycenaean:I9041 3.20
Iran_ChL:I1662 3.10
Levant_N:I1699 2.25
Anatolia_BA:I2495 0.80
Levant_N:I0867 0.75
Mycenaean:I9006 0.65
Minoan_Lasithi:I0074 0.45
Minoan_Odigitria:I9130 0.05
Minoan_Odigitria:I9131 0.05
Iran_ChL:I1674 0.00
Anatolia_BA:I2499 0.00
Anatolia_BA:I2683 0.00
Minoan_Lasithi:I0071 0.00
Minoan_Lasithi:I9005 0.00
Mycenaean:I9010 0.00

Restricted nMonte:

[1] "distance%=0.0281 / distance=0.000281"

Lebanese_Christian:average

Minoan_Lasithi:I0070 32.30
Iran_ChL:I1661 25.25
Levant_N:I1704 24.35
Iran_ChL:I1665 12.50
Mycenaean:I9041 3.80
Levant_N:I1699 1.45
Iran_ChL:I1662 0.35


I think it's quite obvious, the BA Anatolian stems from Ancient Aegeans, which cluster very close.

EDIT: Had to check again, because earlier I got 20% BA Anatolian for Samaritans, albeit the distance was larger than when modelling Samaritans as Levant_N/CHL_Iran (0.5 vs 1.2):


[1] "distance%=1.2936 / distance=0.012936"

Samaritan:GSM537032

Levant_BA:I1706 79.3
Anatolia_BA:I2495 20.6

I suspected that perhaps Bronze Age Anatolians' ancestors might have got CHL Iranian admixture, perhaps at the same time that Neolithic Levantines got it. So I check to see if there's any admixture of CHL Iranian in BA Anatolians, or perhaps BA Anatolians had some Neolithic Levantines (Laziridis study suggest they had some Neolithic Levantine admixture from some Neolithic Levantines bringing agriculture to Anatolia):

[1] "distance%=0.5334 / distance=0.005334"

Anatolia_BA:average

AnatoliaChL:I1584 71.4
Levant_N:I1699 17.2
Levant_N:I1704 11.4

Yep - about 28% Neolithic Levantine, but zero CHL Iranian, which would refute that BA Anatolians had BA Levantine admixture. It would also why Samaritans would yield some good results when modeled like that. Now let's see if there's any Neolithic Levantine admixture among Mycenaeans:

[1] "distance%=0.1203 / distance=0.001203"

Mycenaean:average

Minoan_Odigitria:I9130 31.4
Minoan_Lasithi:I0070 21.9
Srubnaya_Novoselki:I0232 16.8
AnatoliaChL:I1584 12.4
Minoan_Lasithi:I9005 11.6
Levant_N:I0867 5.9

This would be consistent with Lazirdis conclusions, that Mycenaean got less than 6% Neolithic Levantine.

EDIT2: And just as a reference, I've tried to model Mycenaean with BA Anatolians, because they cluster close:

[1] "distance%=0.0312 / distance=0.000312"

Mycenaean:average

Minoan_Lasithi:I9005 24.70
Minoan_Odigitria:I9130 16.35
Anatolia_BA:I2499 14.05
Srubnaya_Barinovka:I0423 10.40
Minoan_Lasithi:I0071 10.00
Anatolia_BA:I2683 9.75
Minoan_Lasithi:I0070 8.05
Srubnaya_Rozhdestveno:I0235 6.10
Minoan_Lasithi:I0074 0.60


Pretty close results - also, 27% Neolithic Levantine from the 24% BA Anatolian would give ~6.5% Neolithic Levantine, pretty similar to what I got above.

To summarize after all the edits, Neolithic Levantine admixture in BA Anatolians cannot be the reason Mycenaeans cluster so close to Western Jews, because:

1. Neolithic Levantines are not the same as Bronze Age Levantines - they are pretty far off.
2. Mycenaeans only had ~6% Neolithic Levantine, which was communicated via BA Anatolians.
3. Trying to model Samaritans with BA Levant + BA Anatolians produce a further result than just modelling them as BA Levantines. 4. The 1.2% short distance we do see, is because of 30% Neolithic Admixture among BA Anatolians, which still produce pretty close results to modern day Samaritans: basically, 80% BA Levant + 6% Neolithic Levant (which probably just overlaps) + 14% non-related admixture.

Idwaajeden
06-09-2018, 12:20 PM
You do realize you are clueless right? My great mother is 100% Polish from Krakow. This is called genetic pull. When you have someone from Southern Europe who mixed with Eastern Europe, the results get centralized between the two regins.

Sure she can be from Krakow, but not Pole.

Do your family tree maybe you can find some clues.

Erikl86
06-09-2018, 12:28 PM
The other thing people get wrong is they see that Mycenaean samples are closer on Oracles and PCA plots to Ashkenazim/south Italians and assume these groups are predominantly Mycenaean, when what they should be seeing is that those groups all aggregate out to the same place even if their admixture components differ. Mycenaeans did not have as much Levant/Arabian/North African type input as do southern Italians and Ashkenazim but they plot in the same place.

Ashkenazim are also not more Mycenaean than mainland Greeks today, they just look like it on PCAs because of the large amount of Slavic in Greece pulling Greeks away from Mycenaeans.

Again, it is documented that there was mass settlements of ancient Greeks all over S. Italy and Sicily. All modern genetic studies show a lot of Greek admixture in modern Sicilians and S. Italians. Again, I feel like I'm repeating myself:

Mycenaean
Minoan_Lasithi 0.780±0.044
Srubnaya 0.220±0.044

Distance: 0.909333794


vs

Greek
Iran_ChL 0.090±0.071
Mycenaean 0.478±0.103
Slav_Bohemia 0.432±0.077

Distance: 0.461783732


Sicilian_East
Bell_Beaker_Germany 0.222±0.077
England_Roman_outlier 0.210±0.134
Mycenaean 0.567±0.163

Distance: 0.504442682


Sicilian_West
England_Roman_outlier 0.216±0.121
Mycenaean 0.503±0.135
Unetice 0.281±0.056

Distance: 0.808464904

Compare that to the BA Sicilians which I've posted earlier. Also, look at this map:

Also, take a look at these maps, first one of Magna Greacia, second one Italian genetics (divided by majority of genes):

23813

23814

Both from wikipedia. Also, on the matter of Greek ancestry among modern day Sicilians:

"However, Greek genetic legacy is estimated at 37% in Sicily" - from wikipedia's article about Italian genetics.

Erikl86
06-09-2018, 12:33 PM
Using North Italians instead of French and Levant BA or Sidon instead of the Levant N would increase the accuracy on the first run, despite the larger distance the first breakdown makes more sense. Altough Ashkenazi Jews may come out as Myceanean like I largely doubt they owe the majority of their ancestry to Myceaneans, rather their admixture makes them Myceanean like. Take the example of a modern Mexican who cluster with Central Asians on a PCA plot despite having no Central Asian ancestry at all but his Iberian and Native American end him plotting with Mixed Old World populations.

I actually posted on this thread about it before, see my results:


[1] "distance%=2.094 / distance=0.02094"

Ashkenazi_Jew:average

Italian_South:SouthItalian10H 100
French_East:French24178 0
French_East:French24433 0
French_East:French24434 0
Polish: Poland4 0
Polish: Poland5 0
Polish: Polish10H 0
Polish: Polish7H 0
Polish: Polish8H 0
Italian_Sicilian_East:EastSicilian1H 0
Italian_Sicilian_East:EastSicilian4H 0
Italian_Sicilian_East:EastSicilian9H 0
Italian_Sicilian_West:WestSicilian10H 0
Italian_Sicilian_West:WestSicilian7H 0
Italian_Sicilian_West:WestSicilian8H 0
Italian_South:SouthItalian16H 0
Italian_South:SouthItalian3H 0

Obviously, bogus results. This proves my assumption that the only reason Sicilians and Italians get clustered so close to Jews, is because of coincidence, or they would be replaceable as the south European component for Ashkenazi Jews, like Cypriots are. It's just that substantial Aegean ancestry (as in South Italians and Sicilians) + Levantine ancestry (from Phoenician, Punic and Arab admixture in S. Italians and Sicilians) + North European ancestry (from Normans mostly and to some very small degree - Lombard/Gothic invasions in the case of Italians) gives a very similar model to Jews.

Now, let's try with similar model to what I've used for Sephardic Jews. It should be basically similar, because both originate from the same "proto" Western Jewish population.

So, just adding Eastern European populations give some close results:

[1] "distance%=0.2075 / distance=0.002075"

Ashkenazi_Jew:average

Cypriot:Cyprus21AJ19 47.8
Cypriot:Cyprus22AJ19 30.4
French_South:SouthFrench3947 14.6
Ukrainian_West:UkrLv240 6.5
Polish: Polish8H 0.7

But, northern European admixture is a little bit too high (plus, Ukrainian should be lower than Polish, IMO, but I guess Western Ukrainians are heavily admixed with Eastern Poles and should be looked at as very close population).

So, let's try to take out Southern French, and replace them with Eastern French, and let's also add Germans:

[1] "distance%=0.1816 / distance=0.001816"

Ashkenazi_Jew:average

Cypriot:Cyprus22AJ19 65.10
Cypriot:Cyprus21AJ19 17.85
German:Germany3 11.60
Ukrainian_West:UkrLv240 5.45

Seems better, plus the amount of Northern European admixture makes much more sense. Notice how Germans take precedence over East French. This is ok - it probably means the original Rhineland inhabitants were more Germanic back then.

Next, let's try and add Samaritans to the exact same mix:

[1] "distance%=0.1964 / distance=0.001964"

Ashkenazi_Jew:average

Samaritan:GSM537032 34.4
Cypriot:Cyprus22AJ19 31.1
French_East:French24434 12.2
Ukrainian_West:UkrLv240 10.3
German:Germany3 6.2
Cypriot:Cyprus21AJ19 5.9

Again, just as in the case of Sephardic Jews, adding Samaritans completely screws up the results, while still retaining a pretty good distance. According to this, Ashkenazi Jews are almost 29% Northern Europeans ! Which is of course wrong. By the way, again see how Samaritans, which represent the closest population to Bronze Age Levantines, never get higher than 30% - just like in my model with ancient Aegeans.

I've even tried to take out the Germans, to see if it'll sort out this problem, while still retaining the Samaritans:

[1] "distance%=0.197 / distance=0.00197"

Ashkenazi_Jew:average

Cypriot:Cyprus22AJ19 33.90
Samaritan:GSM537032 32.75
French_East:French24434 16.05
Ukrainian_West:UkrLv240 11.55
Cypriot:Cyprus21AJ19 5.15
Polish: Polish8H 0.60

Still, same bogus results.

Next, I've took out the Samaritan populations, and tried to put Northern Italians. I've also returned East French. Surely, if Italian is the most significant southern European component, as many here claim, they would grab their share of percentage from Cypriots. Let's see:

[1] "distance%=0.2018 / distance=0.002018"

Ashkenazi_Jew:average

Cypriot:Cyprus21AJ19 39.5
Cypriot:Cyprus22AJ19 37.5
Italian_Bergamo:HGDP01177 9.8
Ukrainian_West:UkrLv240 9.4
French_South:SouthFrench3947 3.9


Well well - Cypriots "took a hit", but is still around 77% - haven't changed a bit. But Bergamo got "only" ~10%, and on the expense of.... East French. Which completely disappeared. This would agree with the fact that Northern Italians have substantial Celtic and even some Germanic make up - like East French. Also, southern, perhaps Greek-derived admixture from S. Italy in Bergamo Italians probably what lowered by 2-3% the percentage Cypriots got.

Tz85
06-09-2018, 12:46 PM
Sure she can be from Krakow, but not Pole.

Do your family tree maybe you can find some clues.

My maternal grandfathers surnames include Zemska, Wesolowski, Rachwal, Gurske, and Elczyk, all of which are Polish. Once again, you are clueless about genetics.

Idwaajeden
06-09-2018, 12:53 PM
My maternal grandfathers surnames include Zemska, Wesolowski, Rachwal, Gurske, and Elczyk, all of which are Polish. Once again, you are clueless about genetics.

Surnames are not the same thing as genetics, there is many Polish and Russian Jews very successful that keep their surname until modern day with -sky, -ski or various suffiexes Polish endings but names are created, genetics is science and history merging.

For fifth time, you are not Polish

Tz85
06-09-2018, 12:59 PM
Surnames are not the same thing as genetics, there is many Polish and Russian Jews very successful that keep their surname until modern day with -sky, -ski or various suffiexes Polish endings but names are created, genetics is science and history merging.

For fifth time, you are not Polish

None of my Polish surnames are Jewish. My maternal grandfathers entire family is from Poland, and have common Polish surnames. You again are clueless. You don't even know how Polish surnames work lol.

Idwaajeden
06-09-2018, 12:59 PM
Here is actual Pole from Krakow on K13, with similar mixed Polish ancestry from western Ukraine as myself


# Population (source) Distance
1 South_Polish 5.01
2 Ukrainian_Lviv 5.85
3 Croatian 6.36
4 Ukrainian 6.41
5 Ukrainian_Belgorod 8.29
6 Polish 8.33
7 Southwest_Russian 8.42
8 Hungarian 8.61
9 Moldavian 8.72
10 Russian_Smolensk 10.05
11 Estonian_Polish 10.93
12 Belorussian 11.04
13 East_German 11.67
14 Austrian 12.49
15 Kargopol_Russian 13.26
16 Serbian 13.85
17 Southwest_Finnish 13.97
18 Estonian 14.75
19 Finnish 14.96
20 Erzya 15.13

Mixed Mode Population Sharing:

# Primary Population (source) Secondary Population (source) Distance
1 90.6% South_Polish + 9.4% Lezgin @ 2.38
2 92.7% South_Polish + 7.3% Abhkasian @ 2.43
3 90.5% South_Polish + 9.5% Kumyk @ 2.44
4 92.1% South_Polish + 7.9% Kurdish @ 2.44
5 92.4% South_Polish + 7.6% Georgian @ 2.45
6 51.1% Ukrainian_Belgorod + 48.9% Hungarian @ 2.47
7 91.3% South_Polish + 8.7% Adygei @ 2.51
8 92.4% South_Polish + 7.6% Georgian_Jewish @ 2.52
9 92.5% South_Polish + 7.5% Armenian @ 2.53
10 91.5% South_Polish + 8.5% Azeri @ 2.56
11 90.7% South_Polish + 9.3% Chechen @ 2.57
12 90.5% South_Polish + 9.5% Tabassaran @ 2.58
13 92.2% South_Polish + 7.8% Iranian @ 2.6
14 91.5% South_Polish + 8.5% North_Ossetian @ 2.6
15 92.6% South_Polish + 7.4% Assyrian @ 2.63
16 91.4% South_Polish + 8.6% Balkar @ 2.7
17 91.1% South_Polish + 8.9% Kabardin @ 2.73
18 92.8% South_Polish + 7.2% Iranian_Jewish @ 2.74
19 91.4% South_Polish + 8.6% Turkish @ 2.74
20 92% South_Polish + 8% Ossetian @ 2.75

Using 2 populations approximation:
1 50% Hungarian +50% Ukrainian_Belgorod @ 2.784264


Using 3 populations approximation:
1 50% Hungarian +25% Ukrainian_Belgorod +25% Ukrainian_Belgorod @ 2.784264


Using 4 populations approximation:
++++++++++++++++++++++++++++++++++++++++++++++++++ +++++++++++++++++++++++++++++++++++++++++++++++
1 Hungarian + Hungarian + Ukrainian_Belgorod + Ukrainian_Belgorod @ 2.784264
2 Bulgarian + North_Swedish + Ukrainian_Belgorod + Ukrainian_Belgorod @ 2.885145
3 Hungarian + Hungarian + Southwest_Russian + Ukrainian_Belgorod @ 3.126389
4 Hungarian + Serbian + Ukrainian_Belgorod + Ukrainian_Belgorod @ 3.186263
5 Bulgarian + North_Swedish + Southwest_Russian + Ukrainian_Belgorod @ 3.272301
6 Bulgarian + Southwest_Finnish + Ukrainian_Belgorod + Ukrainian_Belgorod @ 3.306985
7 North_Swedish + Romanian + Ukrainian_Belgorod + Ukrainian_Belgorod @ 3.329809
8 Austrian + Moldavian + Ukrainian_Belgorod + Ukrainian_Belgorod @ 3.391067
9 Hungarian + Moldavian + Ukrainian_Belgorod + Ukrainian_Belgorod @ 3.405461
10 Serbian + South_Polish + South_Polish + Ukrainian_Belgorod @ 3.411342
11 Croatian + Hungarian + Ukrainian_Belgorod + Ukrainian_Belgorod @ 3.428878
12 East_German + Serbian + Ukrainian_Belgorod + Ukrainian_Belgorod @ 3.442990
13 Hungarian + Lithuanian + Romanian + Ukrainian_Belgorod @ 3.455096
14 Serbian + South_Polish + Ukrainian_Belgorod + Ukrainian_Belgorod @ 3.460484
15 Romanian + South_Polish + Ukrainian_Belgorod + Ukrainian_Belgorod @ 3.461776
16 Romanian + South_Polish + South_Polish + Ukrainian_Belgorod @ 3.462400
17 Austrian + Serbian + Ukrainian_Belgorod + Ukrainian_Belgorod @ 3.471505
18 Hungarian + Serbian + Southwest_Russian + Ukrainian_Belgorod @ 3.498047
19 Romanian + Southwest_Finnish + Ukrainian_Belgorod + Ukrainian_Belgorod @ 3.506355
20 Austrian + Hungarian + Ukrainian_Belgorod + Ukrainian_Belgorod @ 3.527452

Tz85
06-09-2018, 01:09 PM
Here is actual Pole from Krakow on K13, with similar mixed Polish ancestry from western Ukraine as myself


# Population (source) Distance
1 South_Polish 5.01
2 Ukrainian_Lviv 5.85
3 Croatian 6.36
4 Ukrainian 6.41
5 Ukrainian_Belgorod 8.29
6 Polish 8.33
7 Southwest_Russian 8.42
8 Hungarian 8.61
9 Moldavian 8.72
10 Russian_Smolensk 10.05
11 Estonian_Polish 10.93
12 Belorussian 11.04
13 East_German 11.67
14 Austrian 12.49
15 Kargopol_Russian 13.26
16 Serbian 13.85
17 Southwest_Finnish 13.97
18 Estonian 14.75
19 Finnish 14.96
20 Erzya 15.13

Mixed Mode Population Sharing:

# Primary Population (source) Secondary Population (source) Distance
1 90.6% South_Polish + 9.4% Lezgin @ 2.38
2 92.7% South_Polish + 7.3% Abhkasian @ 2.43
3 90.5% South_Polish + 9.5% Kumyk @ 2.44
4 92.1% South_Polish + 7.9% Kurdish @ 2.44
5 92.4% South_Polish + 7.6% Georgian @ 2.45
6 51.1% Ukrainian_Belgorod + 48.9% Hungarian @ 2.47
7 91.3% South_Polish + 8.7% Adygei @ 2.51
8 92.4% South_Polish + 7.6% Georgian_Jewish @ 2.52
9 92.5% South_Polish + 7.5% Armenian @ 2.53
10 91.5% South_Polish + 8.5% Azeri @ 2.56
11 90.7% South_Polish + 9.3% Chechen @ 2.57
12 90.5% South_Polish + 9.5% Tabassaran @ 2.58
13 92.2% South_Polish + 7.8% Iranian @ 2.6
14 91.5% South_Polish + 8.5% North_Ossetian @ 2.6
15 92.6% South_Polish + 7.4% Assyrian @ 2.63
16 91.4% South_Polish + 8.6% Balkar @ 2.7
17 91.1% South_Polish + 8.9% Kabardin @ 2.73
18 92.8% South_Polish + 7.2% Iranian_Jewish @ 2.74
19 91.4% South_Polish + 8.6% Turkish @ 2.74
20 92% South_Polish + 8% Ossetian @ 2.75

Using 2 populations approximation:
1 50% Hungarian +50% Ukrainian_Belgorod @ 2.784264


Using 3 populations approximation:
1 50% Hungarian +25% Ukrainian_Belgorod +25% Ukrainian_Belgorod @ 2.784264


Using 4 populations approximation:
++++++++++++++++++++++++++++++++++++++++++++++++++ +++++++++++++++++++++++++++++++++++++++++++++++
1 Hungarian + Hungarian + Ukrainian_Belgorod + Ukrainian_Belgorod @ 2.784264
2 Bulgarian + North_Swedish + Ukrainian_Belgorod + Ukrainian_Belgorod @ 2.885145
3 Hungarian + Hungarian + Southwest_Russian + Ukrainian_Belgorod @ 3.126389
4 Hungarian + Serbian + Ukrainian_Belgorod + Ukrainian_Belgorod @ 3.186263
5 Bulgarian + North_Swedish + Southwest_Russian + Ukrainian_Belgorod @ 3.272301
6 Bulgarian + Southwest_Finnish + Ukrainian_Belgorod + Ukrainian_Belgorod @ 3.306985
7 North_Swedish + Romanian + Ukrainian_Belgorod + Ukrainian_Belgorod @ 3.329809
8 Austrian + Moldavian + Ukrainian_Belgorod + Ukrainian_Belgorod @ 3.391067
9 Hungarian + Moldavian + Ukrainian_Belgorod + Ukrainian_Belgorod @ 3.405461
10 Serbian + South_Polish + South_Polish + Ukrainian_Belgorod @ 3.411342
11 Croatian + Hungarian + Ukrainian_Belgorod + Ukrainian_Belgorod @ 3.428878
12 East_German + Serbian + Ukrainian_Belgorod + Ukrainian_Belgorod @ 3.442990
13 Hungarian + Lithuanian + Romanian + Ukrainian_Belgorod @ 3.455096
14 Serbian + South_Polish + Ukrainian_Belgorod + Ukrainian_Belgorod @ 3.460484
15 Romanian + South_Polish + Ukrainian_Belgorod + Ukrainian_Belgorod @ 3.461776
16 Romanian + South_Polish + South_Polish + Ukrainian_Belgorod @ 3.462400
17 Austrian + Serbian + Ukrainian_Belgorod + Ukrainian_Belgorod @ 3.471505
18 Hungarian + Serbian + Southwest_Russian + Ukrainian_Belgorod @ 3.498047
19 Romanian + Southwest_Finnish + Ukrainian_Belgorod + Ukrainian_Belgorod @ 3.506355
20 Austrian + Hungarian + Ukrainian_Belgorod + Ukrainian_Belgorod @ 3.527452

Well look at that. Bulgar, Romanian, Serbian, Hungarian, Croatian in the whole list. Thanks for proving to everyone you have no clue what you're talking about, and destroying your own argument.

Idwaajeden
06-09-2018, 01:17 PM
Well look at that. Bulgar, Romanian, Serbian, Hungarian, Croatian in the whole list. Thanks for proving to everyone you have no clue what you're talking about, and destroying your own argument.

This individual here can I trust for a good lasagna sure, but for pierogi I wouldn't trust.

1 Algerian_Jewish + Bulgarian + East_Sicilian + La_Brana-1 @ 5.150026

Polish=

1 Hungarian + Hungarian + Ukrainian_Belgorod + Ukrainian_Belgorod @ 2.784264

giorgio
06-09-2018, 01:17 PM
Both from wikipedia. Also, on the matter of Greek ancestry among modern day Sicilians:

"However, Greek genetic legacy is estimated at 37% in Sicily" - from wikipedia's article about Italian genetics.

That wikipedia source is bogus. The 37% Greek estimate was measured based on the Y-dna, not in regard of what the Sicilian autosomal profile looks like. It's impossible to measure the actual Greek ancestry in Sicilians because generally there's no sharp difference between the Neolithic-Bronze age inhabitans of Sicily and Greece it seems both populations were colonized by the same or genetically similar people.

Idwaajeden
06-09-2018, 01:19 PM
Well look at that. Bulgar, Romanian, Serbian, Hungarian, Croatian in the whole list. Thanks for proving to everyone you have no clue what you're talking about, and destroying your own argument.

Bulgar and Bulgarian is not same thing.

Tz85
06-09-2018, 01:22 PM
This individual here can I trust for a good lasagna sure, but for pierogi I wouldn't trust.

1 Algerian_Jewish + Bulgarian + East_Sicilian + La_Brana-1 @ 5.150026

Polish=

1 Hungarian + Hungarian + Ukrainian_Belgorod + Ukrainian_Belgorod @ 2.784264

Apparently you don't know anything about the Slavic people. Keep digging that hole. You just quoted a Pole, and showed no Pole at 4 populations.

Eihwaz
06-09-2018, 01:29 PM
Apparently you don't know anything about the Slavic people. Keep digging that hole. You just quoted a Pole, and showed no Pole at 4 populations.

I'm not sure why this guy is trying to vehemently claim that this person you're quoting isn't half Polish - Moldavian, rather.

Tz85, your ancestor is half Sicillian half Polish, right? Having Brana and Bulgarian in the Oracle actually sounds about right, though you'd obviously need something significantly more southern to balance that all out to approximate a modern Pole. Thus, the Oracle makes sense to me.

Saying Tz85's ancestor isn't half Polish after looking at those Oracle results would be like saying that my great-grandmother wasn't Central-Eastern European from looking at the results of her half-Ashkenazi child and saying "wow, look at all that southern Euro!!".

Tz85
06-09-2018, 01:33 PM
I'm not sure why this guy is trying to vehemently claim that this person you're quoting isn't half Polish - Moldavian, rather.

Tz85, your ancestor is half Sicillian half Polish, right? Having Brana and Bulgarian in the Oracle actually sounds about right, though you'd obviously need something significantly more southern to balance that all out to approximate a modern Pole. Thus, the Oracle makes sense to me.

Indeed. Unfortunately this person that keeps telling me I'm not Polish, despite my maternal grandfathers entire family is from Poland, and have old Polish surnames, knows absolutely nothing about genetics, or genetic pull.

Erikl86
06-09-2018, 01:34 PM
That wikipedia source is bogus. The 37% Greek estimate was measured based on the Y-dna, not in regard of what the Sicilian autosomal profile looks like. It's impossible to measure the actual Greek ancestry in Sicilians because generally there's no sharp difference between the Neolithic-Bronze age inhabitans of Sicily and Greece it seems both populations were colonized by the same or genetically similar people.

I've already shown it's not the case, Myceneans and BA Sicilians were different, and the fact the modern Sicilians can be modeled as BA Sicilians + Mycenaeans, along with vast linguistic, historic and archaeological records of the mass settlement of Sicily and S. Italy by Iron Age Greeks, pretty much seals the deal that there is substantial Greek admixture among modern day Sicilians and S. Italians:



[1] distance%=5.0349 / distance=0.050349

Beaker_Sicily:I4930

Peloponnese_N 60.55
Peloponnese_N_outlier:I3920 17.6
Armenia_EBA 15.3
WHG 6.25
Moroccan 0.3

[1] distance%=2.3039 / distance=0.023039

Sicilian_East

Peloponnese_N_outlier:I3920 47.9
Yamnaya_Samara 17.8
Armenia_EBA 9.8
Levant_N 8.25
Barcin_N 6.55
Moroccan 6.4
WHG 3.3

[1] distance%=2.1998 / distance=0.021998

Sicilian_West

Peloponnese_N_outlier:I3920 52.9
Yamnaya_Samara 17.05
Moroccan 8.55
Levant_N 8.25
Armenia_EBA 7
WHG 6.25

The Yamnaya admixture is from Mycenaean Greeks, which can be modeled as Peloponnese_N + Steppes people (Yamnaya or Srubnaya) + BA Anatolians:

[1] distance%=1.9802

Mycenaean

Peloponnese_N,45.2
Anatolia_BA,35.8
Yamnaya_Bulgaria,16.4
Balkans_BA,2.6
Hungary_BA,0

giorgio
06-09-2018, 01:38 PM
I've already shown it's not the case, Myceneans and BA Sicilians were different, and the fact the modern Sicilians can be modeled as BA Sicilians + Mycenaeans, along with vast linguistic, historic and archaeological records of the mass settlement of Sicily and S. Italy by Iron Age Greeks, pretty much seals the deal that there is substantial Greek admixture among modern day Sicilians and S. Italians:





Measuring Greek admixture in Sicilians is like measuring Syrian admixture in Lebanese. Makes not much sense given both populations were colonized by Sea peoples and share largely the same DNA with little differences.

Erikl86
06-09-2018, 01:56 PM
Measuring Greek admixture in Sicilians is like measuring Syrian admixture in Lebanese. Makes not much sense given both populations were colonized by Sea peoples and share largely the same DNA with little differences.

I don't know how dominant was contribution from those Sea peoples, but a known, recorded, vast linguistic, ethnic and archaeological evidence for Greek settlement in Sicily is a known fact and scientific consensus - I don't know why we argue about this. The important thing is, that BA Sicilians were not like BA Mycenaean, but today modern Sicilians are much more similar to BA Mycenaeans. That would suggest, again, that a vast ancient Greek settlement had happened.

Idwaajeden
06-09-2018, 02:09 PM
I'm not sure why this guy is trying to vehemently claim that this person you're quoting isn't half Polish - Moldavian, rather.

Tz85, your ancestor is half Sicillian half Polish, right? Having Brana and Bulgarian in the Oracle actually sounds about right, though you'd obviously need something significantly more southern to balance that all out to approximate a modern Pole. Thus, the Oracle makes sense to me.

Saying Tz85's ancestor isn't half Polish after looking at those Oracle results would be like saying that my great-grandmother wasn't Central-Eastern European from looking at the results of her half-Ashkenazi child and saying "wow, look at all that southern Euro!!".

Moldavian is not Polish.

Tz85
06-09-2018, 02:12 PM
Moldavian is not Polish.

Thanks for sabotaging the thread. While you're at it, post your K13 results. Your not Polish, you're I2, and Germanic. ;)

Tz85
06-09-2018, 02:15 PM
Moldavian is not Polish.

They are both ethnically Slavic, now I know you're a totally clueless.

Idwaajeden
06-09-2018, 02:20 PM
They are both ethnically Slavic, now I know you're a totally clueless.

You are not Slavic.

Eihwaz
06-09-2018, 02:23 PM
You are not Slavic.

Let's say that Tz85's ancestor really did cluster with Moldavians, which I doubt.

Moldavians are extremely close to Carpathian Rusyns, Slovaks, and Hungarians - all harboring significant Slavic ancestry.

Tz85
06-09-2018, 02:23 PM
You are not Slavic.

I'm not Slavic? Here are my K13 results. I'm plenty Slavic. You're a troll, and apparently slow.Admix Results (sorted):

# Population Percent
1 North_Atlantic 31.19
2 Baltic 26.86
3 West_Med 16.76
4 East_Med 11.80
5 West_Asian 8.81
6 Red_Sea 1.98
7 South_Asian 1.09
8 Amerindian 1.08


Finished reading population data. 204 populations found.
13 components mode.

--------------------------------

Least-squares method.

Using 1 population approximation:
1 Serbian @ 5.970414
2 Hungarian @ 6.887444
3 Austrian @ 8.163695
4 Romanian @ 9.923835
5 Moldavian @ 9.971128
6 East_German @ 10.427780
7 Croatian @ 10.641770
8 Bulgarian @ 13.367585
9 West_German @ 13.531508
10 South_Dutch @ 15.215054
11 French @ 15.616653
12 South_Polish @ 17.428213
13 Ukrainian_Lviv @ 17.619133
14 North_German @ 18.950190
15 Ukrainian @ 19.080879
16 North_Italian @ 19.244318
17 Portuguese @ 20.365267
18 Spanish_Galicia @ 20.373585
19 Spanish_Cataluna @ 20.948429
20 Southeast_English @ 21.507055

Using 2 populations approximation:
1 50% Austrian +50% Romanian @ 2.010429


Using 3 populations approximation:
1 50% Croatian +25% Greek_Thessaly +25% Southwest_English @ 1.322883


Using 4 populations approximation:
++++++++++++++++++++++++++++++++++++++++++++++++++ +++++++++++++++++++++++++++++
1 Croatian + Greek_Thessaly + Moldavian + Southwest_English @ 1.101335
2 Bulgarian + East_German + French + Moldavian @ 1.142167
3 Croatian + East_German + East_German + Italian_Abruzzo @ 1.256367
4 Croatian + South_Italian + Southwest_English + Ukrainian @ 1.307890
5 Croatian + Croatian + Greek_Thessaly + Southwest_English @ 1.322883
6 Moldavian + Polish + South_Italian + Southwest_English @ 1.333534
7 Austrian + Bulgarian + French + Moldavian @ 1.340562
8 Croatian + South_Italian + Southwest_English + Ukrainian_Lviv @ 1.346728
9 Croatian + East_German + Greek_Thessaly + West_German @ 1.362655
10 Central_Greek + East_German + South_Dutch + Ukrainian_Lviv @ 1.383185
11 Croatian + Moldavian + North_German + Tuscan @ 1.390033
12 East_German + Greek_Thessaly + Moldavian + South_Dutch @ 1.403677
13 French + Hungarian + Moldavian + Romanian @ 1.404980
14 Bulgarian + Croatian + Romanian + Southwest_English @ 1.417810
15 Central_Greek + East_German + East_German + East_German @ 1.421180
16 Central_Greek + East_German + South_Dutch + Ukrainian @ 1.445671
17 Greek_Thessaly + Polish + Romanian + Southwest_English @ 1.452126
18 East_German + East_German + Moldavian + Tuscan @ 1.462900
19 Austrian + Hungarian + Moldavian + North_Italian @ 1.467752
20 Croatian + Greek_Thessaly + Moldavian + West_Scottish @ 1.471375

Idwaajeden
06-09-2018, 02:26 PM
Let's say that Tz85's ancestor really did cluster with Moldavians, which I doubt.

Moldavians are extremely close to Carpathian Rusyns, Slovaks, and Hungarians - all harboring significant Slavic ancestry.

Not exactly, Moldavians are different than the groups you mention

Eihwaz
06-09-2018, 02:57 PM
Please look at the G25 and K36 PCAs and then tell me that again.

Tz85
06-09-2018, 03:01 PM
Please look at the G25 and K36 PCAs and then tell me that again.

Remember, this kid isn't very bright

Ajeje Brazorf
06-09-2018, 04:50 PM
The other thing people get wrong is they see that Mycenaean samples are closer on Oracles and PCA plots to Ashkenazim/south Italians and assume these groups are predominantly Mycenaean, when what they should be seeing is that those groups all aggregate out to the same place even if their admixture components differ. Mycenaeans did not have as much Levant/Arabian/North African type input as do southern Italians and Ashkenazim but they plot in the same place.

Ashkenazim are also not more Mycenaean than mainland Greeks today, they just look like it on PCAs because of the large amount of Slavic in Greece pulling Greeks away from Mycenaeans.

Wasn't TheApricity enough to spread your obsessions and nonsense, you do it in other forums too? We are not more Arab/African than the Mycenaeans and even in the PCA those Mycenaeans are distant from Ashkenazi, Jews and any other population.

I9041
1 Algerian_Jewish 11.537820
2 South_Italian 11.728889
3 West_Sicilian 11.930955
4 Italian_Jewish 11.959280
5 Sephardic_Jewish 13.163100
6 East_Sicilian 14.161626
7 Central_Greek 14.995540
8 Italian_Abruzzo 15.194218
9 Tunisian_Jewish 16.531200
10 Tuscan 16.534834
11 Ashkenazi 17.861904
12 Greek 18.451389
13 Libyan_Jewish 19.168669
14 Greek_Thessaly 20.568029
15 Cyprian 20.654701
16 North_Italian 22.197578
17 Tunisian 23.566832
18 Algerian 25.193422
19 Mozabite_Berber 26.101456
20 Lebanese_Muslim 26.503309

I9033
1 East_Sicilian 13.041629
2 South_Italian 13.528161
3 Central_Greek 13.583624
4 Tuscan 14.170655
5 West_Sicilian 14.342082
6 Italian_Abruzzo 14.535425
7 Greek_Thessaly 14.860162
8 Ashkenazi 16.069584
9 Italian_Jewish 16.534094
10 Greek 16.571495
11 Algerian_Jewish 17.784563
12 North_Italian 18.839813
13 Sephardic_Jewish 19.146748
14 Libyan_Jewish 21.675129
15 Tunisian_Jewish 22.617178
16 Bulgarian 23.100882
17 Algerian 24.609713
18 Romanian 25.011593
19 Cyprian 25.206230
20 Tunisian 25.505022

Sikeliot
06-09-2018, 05:44 PM
Why won't it make sense to you? It makes perfect sense - and would also explain, again, why Western Jews cluster so close to Sicilians and S. Italians, and why they get so short distance when modeled with Cypriots.

Geographically it does not make sense but I know it is the case autosomally. I would argue as I have been, though, that the Caucasian and Levantine/Arabian components may have entered Sicily and Cyprus on separate dates, with the Caucasian having arrived earlier and the Levantine/Arabian being more likely from historical settlement from the Near East.



I don't know why you claim that there is little to no IBD sharing between mainland Greece and southern Italy, when there's an article from last proving just the opposite:

https://www.nature.com/articles/ejhg201718

There is no contradiction between what I am saying and what you just posted. The Sicilian samples in that study, like in most academic studies, come from southeastern Sicily (Syracuse, southern Catania area) and Trapani, where there is higher Steppe admixture, higher IBD sharing with Greece, and the largest, most continuous contact with Greece. What I said is that there is little to no IBD sharing between Calabria and the East-Central Sicily region (Palermo, Messina, Agrigento, etc) and mainland Greece, as shown by Sarno et al which I quoted multiple times to you, and these are the regions that deviate furthest from mainland Greeks and Balkan people, are the most firmly in the "eastern Mediterranean" continuum populations which ALL mainland Greek populations fall outside of, and have the highest Near Eastern input.

If you want to see another source backing this up, this study (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4757772/) shows the same, that there is clear evidence of ancient Greek ancestry in eastern Sicily (the samples were taken from various towns in Syracuse and Ragusa provinces, in southeast Sicily, which you can see in an attachment at the end of the study) and an Apulia-based southern Italy sample, but much less in either western Sicily and Ionian Italy (Calabria).

Therefore, it makes complete sense that Peloponnesians will share more with a Sicilian sample based on southeast Sicily, which deviates from the mainstream Sicilian cluster and is closer to Apulia and has higher evidence of mainland Greek input.

Also, the 37% Y-dna contribution of Greece in Sicily is significantly bolstered by a lot of the sampled towns in the study being in the southeast of Sicily (southern part of Enna, Ragusa area). The study even states that Greek haplogroups in western Sicily were low.

"Samples from East Sicily, West Sicily, South and Central Italy showed significant (P<0.01) enrichment of GChps when the Greek sample from the Euboea Island was compared with Corinthia and reference sources. Except in West Sicily and Central Italy, this enrichment remained highly significant even after correcting for multiple tests (P<0.05).41 Conversely, when considering Corinthia against other reference sources, none of the recipient samples showed a full set of significant values. The comparisons involving recipients versus Albania most commonly showed a lack of significant enrichment in GChps with respect to Euboea and Corinthia. We reasoned that contacts either between sources and recipients or between sources after the GC, that is, during the Classical and Christian periods, might have contributed to increase the GChps rate. Thus, to provide more stringent conditions for haplotype identification, we excluded all haplotypes with a molecular distance less than seven mutational steps. This 'filtered' data set confirmed the pattern observed with the less stringent criteria for the East Sicily/Euboea pair, which showed significant enrichment in GChps in two out of three comparisons even after the Bonferroni correction (Table 1b). The results for West Sicily and South Italy did not hold statistical significance when a Bonferroni correction was applied. None of the other Italian recipients showed a full set of significant enrichment with Euboea or Corinthia."


It's not that they don't cluster with S. Italians and Sicilians - it's that they have from 0.2-14.4% Slavic admixture, which make modern Greek Islanders to cluster closer to S. Italians and Sicilians, than mainland Greeks, but mainland Greeks still cluster close to Italians. The more north you go in Greece, the more Slavic admixture you encounter. Greek Islanders are least admixed with Slavic immigrants, while the vast MAJORITY of Greek ancestry in Italy and Sicily predates the arrival of Slavic tribes to Greece, even if later on few Greek mainlanders also arrived to Italy, they were by far a minority. It's quite obvious.

They cluster closely because of shared ancient ancestry, not because of demographic displacement. However, I will add that if there was to be IBD sharing, it'd be more likely because the Byzantines, after expelling some of the Slavs from the Peloponnese, repopulated the Peloponnese with Sicilian and Calabrese people as well as Anatolian Greeks. Thus there is just as much, if not more, south Italian ancestry in mainland Greece than the reverse. Still, the important piece here is that these conclusions are based on southeast Sicilian samples. If they were based on Palermo or Messina instead, the commonalities with mainland Greece would drop off.



That wikipedia source is bogus. The 37% Greek estimate was measured based on the Y-dna, not in regard of what the Sicilian autosomal profile looks like. It's impossible to measure the actual Greek ancestry in Sicilians because generally there's no sharp difference between the Neolithic-Bronze age inhabitans of Sicily and Greece it seems both populations were colonized by the same or genetically similar people.

This is my point, they have shared ancestry but it does not mean that there was a replacement of one population in favor of another. As we see more ancient samples from southern Italy I believe we will continue to see Mycenaean and Minoan-like populations having existed in southern Italy since well before the Greek colonization.


I don't know how dominant was contribution from those Sea peoples, but a known, recorded, vast linguistic, ethnic and archaeological evidence for Greek settlement in Sicily is a known fact and scientific consensus - I don't know why we argue about this. The important thing is, that BA Sicilians were not like BA Mycenaean, but today modern Sicilians are much more similar to BA Mycenaeans. That would suggest, again, that a vast ancient Greek settlement had happened.


No one is denying Greek-speaking people in southern Italy nor even settlement of people from Greece into southern Italy, but I am absolutely denying that there was a large contribution of mainland Greeks into southern Italy, with the exception of Apulia, southeast Sicily, Trapani and other regions that show higher IBD sharing with them. Some of this could also be due to assimilated Arbereshe populations. You do realize even Aegean islanders are predominantly Hellenized natives, right?

So either there was a mass Greek settlement in Sicily but Greeks in mainland Greece today are essentially unrelated to the Greeks of past eras, or the shift was more cultural than genetic.


Wasn't TheApricity enough to spread your obsessions and nonsense, you do it in other forums too? We are not more Arab/African than the Mycenaeans and even in the PCA those Mycenaeans are distant from Ashkenazi, Jews and any other population.

I could easily say the same -- wasn't TA enough for you to have one place to make nasty comments to me, now you have to do it here also? I'm not interested in arguing with you, and I will not reply to any more of your posts neither here nor anywhere else.

alexfritz
06-09-2018, 05:46 PM
I've already shown it's not the case, Myceneans and BA Sicilians were different, and the fact the modern Sicilians can be modeled as BA Sicilians + Mycenaeans, along with vast linguistic, historic and archaeological records of the mass settlement of Sicily and S. Italy by Iron Age Greeks, pretty much seals the deal that there is substantial Greek admixture among modern day Sicilians and S. Italians:



The Yamnaya admixture is from Mycenaean Greeks, which can be modeled as Peloponnese_N + Steppes people (Yamnaya or Srubnaya) + BA Anatolians:

[1] distance%=1.9802

Mycenaean

Peloponnese_N,45.2
Anatolia_BA,35.8
Yamnaya_Bulgaria,16.4
Balkans_BA,2.6
Hungary_BA,0

just a question, if the EBA sicilians were differing from the mycenaean type (bar steppe being one feature) than how close would you say they are to minoan or anatolia_chl type? could the sicilians from 2200-1900bc pose as part of a new substrate pop stretching from east(crete/peleponnese)>west(sicily) super/overseding the earlier cardial/impressed-ware farmers?

an alternate model from Eurogenes I4930 to #85
[1] distance%=4.7903 Beaker_Sicily_no_steppe
Barcin_N,54
Anatolia_ChL,40.2
Blatterhole_HG,5.8

Ajeje Brazorf
06-09-2018, 06:17 PM
I could easily say the same -- wasn't TA enough for you to have one place to make nasty comments to me, now you have to do it here also? I'm not interested in arguing with you, and I will not reply to any more of your posts neither here nor anywhere else.

Neither am I interested in discussing with you, since it would be useless and you would always continue with your obsessions as you have done in recent years.

Claudio
06-09-2018, 06:46 PM
Yep. Also worth mentioning that there were Greek-speaking Romaniote community in S. Italy - and not the other way around (there was no documented Italkim communities in the Eastern Mediterranean). See: https://en.wikipedia.org/wiki/Italian_Jews#Graeco-Italian_Jews

There were two genetic bottlenecks experienced by Ashkenazi Jews.
Do you think they may influence the outcome of your Admixture calculations?
Inbetween the first bottleneck and the second bottlenecks could there have been Admixture received from Calabrian Jews?
23822
23823
23824
23825
23826

Sikeliot
06-09-2018, 07:07 PM
What about the possibility that Italkim (Italian Jews) might be the remnants in Italy of an ancestral group to both Ashkenazim and Sephardim?

giorgio
06-09-2018, 07:14 PM
We need old Jewish samples possibly from Roman era cemeteries to estimate how they scored genetically back then. I assume they'd already had picked up some Hellenic-Roman ancestry but still were mainly Levant like. (Another possibility and this would be the the alternative that the Mycenaean affinity already existed in Judea by absorbing Philistine ancestry)

Sikeliot
06-09-2018, 07:23 PM
We need old Jewish samples possibly from Roman era cemeteries to estimate how they scored genetically back then. I assume they'd already had picked up some Hellenic-Roman ancestry but still were mainly Levant like. (Another possibility and this would be the the alternative that the Mycenaean affinity already existed in Judea by absorbing Philistine ancestry)

What I take issue with is that we keep calling it "Mycenaean" as if we are trying to turn everyone from Sicilians to Jews to Levantine Arabs into Greeks. Mycenaeans are simply a proxy for what might have been a wave of migration out of Anatolia and might have had nothing to do with Philistines, Mycenaeans, etc. at all.

This is the same way we could call ENF ancestry "Sardinian admixture" but it isn't, simply Sardinians best proxy it. For Mycenaean-like admixture it's best proxied by south Italy, because mainland Greeks have seen up to 30% of their ancient DNA replaced by Slavs.

Erikl86
06-09-2018, 07:34 PM
To everyone.... I've just re-read Hebar et al. article from 2017... check out the following, it has been there all along:


Although f4 tests showed that present-day Lebanese share significantly more alleles with Sidon_BA than other Near Eastern populations do, indicating genetic continuity, we failed to model the present-day Lebanese using streams of ancestry coming only from Levant_N and Iran_ChL (qpAdm rank1 p = 8.36 × 10−7), in contrast to our success with Sidon_BA. We therefore further explored our dataset by running ADMIXTURE54 in a supervised mode using Western hunter-gatherers (WHG), Eastern hunter-gatherers (EHG), Levant_N, and Iran_N as reference populations. These four populations have been previously13 found to contribute genetically to most West Eurasians. The ADMIXTURE results replicate the findings from qpAdm for Sidon_BA and show mixture of Levant_N and ancient Iranian populations (Figure 3A). However, the present-day Lebanese, in addition to their Levant_N and ancient Iranian ancestry, have a component (11%–22%) related to EHG and Steppe populations not found in Bronze Age populations (Figure 3A). We confirm the presence of this ancestry in the Lebanese by testing f4(Sidon_BA, Lebanese; Ancient Eurasian, Chimpanzee) and find that Eurasian hunter-gatherers and Steppe populations share more alleles with the Lebanese than with Sidon_BA (Figures 3B and S14). We next tested a model of the present-day Lebanese as a mixture of Sidon_BA and any other ancient Eurasian population using qpAdm. We found that the Lebanese can be best modeled as Sidon_BA 93% ± 1.6% and a Steppe Bronze Age population 7% ± 1.6% (Figure 3C; Table S6). To estimate the time when the Steppe ancestry penetrated the Levant, we used, as above, LD-based inference and set the Lebanese as admixed test population with Natufians, Levant_N, Sidon_BA, Steppe_EMBA, and Steppe_MLBA as reference populations. We found support (p = 0.00017) for a mixture between Sidon_BA and Steppe_EMBA which has occurred around 2,950 ± 790 ya (Figure S13B). It is important to note here that Bronze Age Steppe populations used in the model need not be the actual ancestral mixing populations, and the admixture could have involved a population which was itself admixed with a Steppe-like ancestry population. The time period of this mixture overlaps with the decline of the Egyptian empire and its domination over the Levant, leading some of the coastal cities to thrive, including Sidon and Tyre, which established at this time a successful maritime trade network throughout the Mediterranean. The decline in Egypt’s power was also followed by a succession of conquests of the region by distant populations such as the Assyrians, Persians, and Macedonians, any or all of whom could have carried the Steppe-like ancestry observed here in the Levant after the Bronze Age.

Being that Mycenaeans have been found to be best modeled as ~80% Minoans + ~20% Steppes-like people, according to Steppe admixture in Mycenaeans, lots of Caucasus admixture already in Minoans (Lazaridis et al. 2017) (https://www.nature.com/articles/nature23310):

However, the Mycenaeans differed from Minoans in deriving additional ancestry from an ultimate source related to the hunter–gatherers of eastern Europe and Siberia [6, 7, 8], introduced via a proximal source related to the inhabitants of either the Eurasian steppe [1, 6, 9] or Armenia [4, 9]. Modern Greeks resemble the Mycenaeans

Or:

Mycenaean
Minoan_Lasithi 0.786±0.049
Sintashta 0.214±0.049
Distance: 0.96574059

Mycenaean
Corded_Ware_Germany 0.210±0.043
Minoan_Lasithi 0.790±0.043
Distance: 0.961238695

Mycenaean
Minoan_Lasithi 0.791±0.043
Srubnaya 0.209±0.043
Distance: 0.950419642

And according to what I've found, as explained in the beginning:

[1] "distance%=0.0275 / distance=0.000275"

Lebanese_Christian:average

Minoan_Lasithi:I0070 34.70
Iran_ChL:I1661 25.90
Levant_N:I1704 22.30
Iran_ChL:I1665 11.35
Levant_N:I0867 1.85
Mycenaean:I9041 1.60
Minoan_Lasithi:I0073 1.20
Iran_ChL:I1662 1.05
Minoan_Lasithi:I0074 0.05


Or, if I just use Myceneaens:

[1] "distance%=0.082 / distance=0.00082"

Lebanese_Christian:average

Levant_N:I1704 34.00
Mycenaean:I9041 24.45
Iran_ChL:I1661 23.50
Iran_ChL:I1665 9.45
Levant_N:I1699 8.60

It could very well be, as I said before, that nMonte finds it difficult to differetiates between Minoans and Mycenaeans, because they are highly similar, or because Neolithic Levantine overlaps with the small amount of Neolithic Levantine found in Mycenaeans, and so it filters out and decides Minoans are a better fit rather than Mycenaean.

Anyhow, 20% Steppes-like admixture from 24.5% of Mycenaean = ~5%, pretty similar to the 7% Steppes-like admixture Haber et al. found !

And he himself suggests that "the admixture could have involved a population which was itself admixed with a Steppe-like ancestry population.".

He also proposes Macedonian - that is, Greek people which would have been very similar to Mycenaeans genetically, from the Hellenistic period - origins as one of the sources.

The timing is also exactly right - 2,950 ± 790 ya, encompassing Phoenician start of trading accross the Mediterranean, and ofcourse if there is an error of ± 790 years then it covers Hellenistic period.

Seems like my own theory gets re-enforcement from the last two relevant papers .

Claudio
06-09-2018, 07:39 PM
Regarding scientific study findings that Ashkenazi Jews did not share dna with Italians till much later than expected.
I found this Article on some page on the web.
Some of it was pure cod shit. Lol
But some of the things were worthy of discussion so I have tried to upload info in next two posts.
23827
23828
23829
23830
23831

Claudio
06-09-2018, 07:44 PM
Continued..
23832
23833
23834
23835
Opinions?

Sikeliot
06-09-2018, 08:07 PM
He also proposes Macedonian - that is, Greek people which would have been very similar to Mycenaeans genetically, from the Hellenistic period - origins as one of the sources.

There is no reason to believe Macedonians would have been similar to the Mycenaean samples. Macedonians would have been closer to the Thracian population and likely would have had much more Steppe than them. Also, it is unlikely, during the era that Macedonians conquered the Near East, that there were still people in Greece who resembled Mycenaeans genetically... Mycenaeans were long admixed with other populations.

Erikl86
06-09-2018, 08:43 PM
There is no reason to believe Macedonians would have been similar to the Mycenaean samples. Macedonians would have been closer to the Thracian population and likely would have had much more Steppe than them. Also, it is unlikely, during the era that Macedonians conquered the Near East, that there were still people in Greece who resembled Mycenaeans genetically... Mycenaeans were long admixed with other populations.

When people say "Macedonia", as in the ancient Macedonian Empire, they don't mean specifically Greeks from Macedonia. Before they expanded east, they've united all of Greece. It is very well be that the majority of Greeks that would arrive to the Levant would be from close-by areas, such as Greek Islands, simply for geographic simplicity. Also, Mycenaean admixture shows up in high percentage in modern-day Greeks - no reason to think that Greeks from 2400 years ago, well before Greeks have brought under their reign via set of multi-centuries, inter-continental empires (Hellenistic, then Byzantine) millions of people from other ethnic backgrounds, and well before the major Slavic immigration to the Balkan, that they would be less related to them. It would be only 600 years after Mycenaean civilization finally transitioned into Classical Greek one. And the Greeks themselves wrote at those times how they were the descendants of the Mycenaeans.

Check out qpAdm models of modern mainland Greeks vs. Mycenaeans:


Mycenaean
Minoan_Lasithi 0.780±0.044
Srubnaya 0.220±0.044
P-value 0.909333794

chisq 7.595
Full output

vs

Greek
Iran_ChL 0.090±0.071
Mycenaean 0.478±0.103
Slav_Bohemia 0.432±0.077

P-value 0.461783732
chisq 12.820

Even today, Mycenaean admixture gets almost 50% in modern highly Slavic-mixed mainland Greeks.

Erikl86
06-09-2018, 08:52 PM
Opinions?

I've actually addressed this study in the opening post of this thread:



There is one issue which I've yet managed to work around, and is the timing of when the admixture with southern Europeans happened.

According to a study from 2016, the timing of the admixture with southern Europeans, mainly southern Italians, between 35-55 generations ago (500-1000 AD), and this their conclusions:

What is perhaps surprising is the timing of the Southern European admixture to ≈31-52 generations
ago, since Jews are known to have resided in Italy already since antiquity. This result would then imply
no gene flow between Jews and local Italian populations until the turn of the millennium, either due to
endogamy, or because the group that eventually gave rise to contemporary Ashkenazi Jews did not
reside in Southern Europe until that time. More detailed or alternative interpretations are left for future
studies.
What if the solution is not that the Jews didn't mix with local Italians, but that until 31-52 generations ago, there was little Greek gene flow into the rest of southern Italians?

Until 640-750 AD, most southern Italy and Sicily were still controlled by Greek speaking Byzantine Empire, and Greek was still spoken by the majority of people in some southern areas in Italy and Sicily all the way to the 11th century.

After the Lombards conquered the rest of Italy, and then were absorbed by the Frankish empire in the 9th century, the area began a process of Latinization, which while was occurring before in small amounts, accelerated immensely after Southern Italy and Sicily were completely disconnected from the Byzantine Empire.

That would mean those Greek people were then absorbed into general Latin-speaking population. But that was only the start of a gradual absorption of Italian-Greeks into the rest of the population - after the Muslim conquest of Sicily and continuous raids by Muslims on continental Italy, many Christians (of mainly Greek ancestry, as was the case for southern Italy and Sicily) fled to mainland Italy and further north.

Sicily itself took more than 3 decades to be conquered by Muslims.

I believe this integration and movement of southern, mainly Greek background, Italians with the rest of non-Greek Italians, actually took place in that time frame - 500-1000 AD. Before that, while of course there was some exchange of population, Greek was an official language of the Roman Empire (even after the split, it was still official in both halves of the empire) and so Greek speaking people could have kept their identity as such and had no pressure at all to speak any other language (all official Roman positions dictated that you'd know both Greek and Latin).

Perhaps then, "This result would then imply no gene flow between Jews and local Italian populations until the turn of the millennium, either due to endogamy, or because the group that eventually gave rise to contemporary Ashkenazi Jews did not reside in Southern Europe until that time." actually means that Greek and Phoenician admixture with Italians didn't occur in large amounts until the turn of the 1st millennium AD, thus creating the illusion that Jews didn't mix with local population there - they did, it's just that their European admixture was Mycenaean-Greek, and that admixture didn't occur among Italians until after the absorption of the Magnea Grecia by Latin-Italy around the turn of the millennium.

Basically, my opinion is that it's not that Ashkenazim mixed with Italians only after 500 AD, it's that Italians have begun absorbing Greeks from S. Italy and Sicily en-mass starting that time, and being that Ashkenazim, according to my suggestion, are about 50% admixed with Aegeans, this is when a very similar admixture from non-Jewish Greco-Italians would start appearing among the rest of the Italian population. The time period fits when Byzantines lost for good big parts of Greek S. Italy, a process which followed with their Latinization and absorption in Latin-speaking Italy, and ended mostly around the turn of the millennium, with the Byzantines losing all of their territories in the Italian peninsula. The time frame fits, it seems.

Sikeliot
06-09-2018, 09:09 PM
Check out qpAdm models of modern mainland Greeks vs. Mycenaeans:



Even today, Mycenaean admixture gets almost 50% in modern highly Slavic-mixed mainland Greeks.

Based on this, Greeks are almost half Mycenaean, half Central European Slavic, with about 9% additional Iranian/CHG. This is quite a large population replacement.

Sikeliot
06-09-2018, 09:10 PM
Basically, my opinion is that it's not that Ashkenazim mixed with Italians only after 500 AD, it's that Italians have begun absorbing Greeks from S. Italy and Sicily en-mass starting that time, and being that Ashkenazim, according to my suggestion, are about 50% admixed with Aegeans, this is when a very similar admixture from non-Jewish Greco-Italians would start appearing among the rest of the Italian population. The time period fits when Byzantines lost for good big parts of Greek S. Italy, a process which followed with their Latinization and absorption in Latin-speaking Italy, and ended mostly around the turn of the century, with the Byzantines losing all of their territories in the Italian peninsula. The time frame fits, it seems.


I don't see why this is your explanation for it rather than that all of them (Aegean islands, Sicily, south Italy, and Ashkenazim) have significant Levantine input.

Claudio
06-09-2018, 10:07 PM
I don't see why this is your explanation for it rather than that all of them (Aegean islands, Sicily, south Italy, and Ashkenazim) have significant Levantine input.

I would imagine and agree that all Aegean Islands,Sicily,South italy And Ashkenazim do indeed have significant Levantine input.
23839
23840
23841
23842
Also mentioned in other articles is that Crete was also occupied by Muslim invaders(control Crete and you control the Eastern Mediterranean.
The Levant was the launching pad for all Muslim navy expansion.
Demographically North African Armies were Mostly Berber.
But The Arab Muslim Armies of the near east though Controlled by Arabians were mostly made up of mercenaries from all over Levant and Caucasus.
The Army was run by a feudal system with the bulk of Arab Army made up from local Sultans/lords of conquered Levant and west Asia inlisting peasants who worked the land.
To control this gigantic rabble the Sultans had mercenaries but more importantly the mamluks.
These were slave children from mostly Syria,Turkey & Iran who were hand picked for being physically strong and healthy who were then taken from there family then trained in the art of war to be crack troops.
On reaching adulthood they were given there freedom but given the choice to stay in the employ of the Sultans which they mostly did as they had emotionally become attrached to there masters.
These warriors became the organization of the Muslim Armies and had the inner confidence of the Sultans.
This was a genius system of control on the Sultans part but which centuries later would backfire in the mamluk uprising in which this slave warrior elite had enough and slit the throats of there masters taking there place and so began the empire of the mamluks.
So ethnically other than the Berber Armies of North Africa the Arab Armies of the Middle East were most likely ethnically more Levantine than actual Arabian.
So for instance sicily’s Population doubled under Muslim Control hundreds of old towns repopulated hundreds of new towns founded.
Some were Berber towns Some where Arab Towns.
But the Arab population of Arab Towns was most likely more Levantine and Caucasus in ancestry than actual Arabian.

Eihwaz
06-10-2018, 12:05 AM
Anyhow, 20% Steppes-like admixture from 24.5% of Mycenaean = ~5%, pretty similar to the 7% Steppes-like admixture Haber et al. found !

Interesting point, though I'd be much more inclined to agree with it if it wasn't for a more likely scenario, which was Steppe ancestry coming in through the Mittani (there's R1a-Z94 in the Levant, pretty Indo-Iranian signal).

giorgio
06-10-2018, 04:00 AM
I don't see why this is your explanation for it rather than that all of them (Aegean islands, Sicily, south Italy, and Ashkenazim) have significant Levantine input.

Ashkenazim have significant Levantine input, the rest mostly derive their "Levantine" ancestry from Bronze age Anatolian populations. Compared to Bronze Age Sicilian samples modern Sicilians have + 10-15% North African and Levantine which is rather low and the extra Steppe balance out the Near Eastern so in other words modern Sicilians became both more Continental European and Near Eastern still don't really differ from Bronze age samples - Mainland Greeks on the other hand received significant Steppe or let's say some Slavic input that changed their position compared to Bronze age Hellenic samples.

giorgio
06-10-2018, 04:03 AM
Based on this, Greeks are almost half Mycenaean, half Central European Slavic, with about 9% additional Iranian/CHG. This is quite a large population replacement.

That is true in the case of some genetic outlier samples from Northern Greece. Peloponnese and Southern Greeks would be more Myceanean and less slavic than that. Islanders and Western Asia Minor Greeks have higher Myceanean than any Mainlander which answer some of your previous questions.

Targum
06-10-2018, 05:24 AM
What about the possibility that Italkim (Italian Jews) might be the remnants in Italy of an ancestral group to both Ashkenazim and Sephardim?

This is obviously true and well discussed on the forum
https://anthrogenica.com/showthread.php?14484-Could-Western-Jews-(Ash-and-Seph-)-descend-from-Aegeans-and-Levantine-admixture&p=411065&viewfull=1#post411065

Erikl86
06-10-2018, 06:18 AM
I don't see why this is your explanation for it rather than that all of them (Aegean islands, Sicily, south Italy, and Ashkenazim) have significant Levantine input.

Because the Levantine admixture among Italians, Aegeans and of course Jews (which were originally a Levantine population) was much earlier than 500 - 1000 AD, and so a different explanation needs to solve this apparent contradiction.

Also, you'd have to explain why Samaritans, which also derive from BA Levantines, seem to lack this Aegean admixture. The only way to explain this, IMO, is that this admixture probably mostly started after Hellenistic period, before this was perhaps limited mostly to southern Levantine shores.

Since both both Palestinian Christians and Lebanese Christians descend probably from a core of Hellenistic Jews which later own became the first Christian communities in the Levant, and then more non-Jewish Hellenized Levantines joined them, then all of these groups - Western Jews, Christian/Muslim Lebanese, Druze and Christian Palestinians should have significant Aegean admixture - which is what I got, and apparently, as I've explained, this has some support from previous finding by Hebar et al. with his own suggestion that BA Levantines admixed with population that had significant Steppes-ancestry.

In all my nMonte runs I've found Druze, Lebanese and Christian Palestinians to be 25-35% admixed with Aegeans, to yield the closest distance.

Samaritans yield none. This would go along with history for several reasons:
1. Just like post Babylonian captivity Jews, Samaritans didn't marry non-Samaritans without conversion.
2. As opposed to Second Temple Jews, Samaritans were not in position of power to force other nations around them to convert, plus there was no proselytizing movement among Samaritans that is known of.

And Samaritan religion and identity was already formed when the Hellenistic period began, so that would make sense they have little, if any, Aegean admixture.

Erikl86
06-10-2018, 06:23 AM
Interesting point, though I'd be much more inclined to agree with it if it wasn't for a more likely scenario, which was Steppe ancestry coming in through the Mittani (there's R1a-Z94 in the Levant, pretty Indo-Iranian signal).

Unlikely, as Hurrians would have brought this earlier. Haber et al. himself suggests this happened 2900 +- ~750 years ago, after the Mitanni were already gone. And I don't know if Hurrians and Mitanni actually had significant Steppes-like admixture.

I still don't understand why you reject the possibility of substantial Greek admixture among Levantines. They have been settling in the Levant since the 10th century BC and Greek became lingua franca at one point in that region along side Aramaic. Also, Judaism itself was very popular among 1st century Greco-Romans.

Erikl86
06-10-2018, 06:39 AM
just a question, if the EBA sicilians were differing from the mycenaean type (bar steppe being one feature) than how close would you say they are to minoan or anatolia_chl type? could the sicilians from 2200-1900bc pose as part of a new substrate pop stretching from east(crete/peleponnese)>west(sicily) super/overseding the earlier cardial/impressed-ware farmers?

an alternate model from Eurogenes I4930 to #85
[1] distance%=4.7903 Beaker_Sicily_no_steppe
Barcin_N,54
Anatolia_ChL,40.2
Blatterhole_HG,5.8

Yes Minoans were related to other Neolithic-derived Europeans, including pre-Greek colonization BA Sicily. I would say the closest living Europeans genetically would be modern-day Sardinians.

Generalissimo
06-10-2018, 07:10 AM
You guys should use the Global25 for this, because it picks up recent drift better than the BR K7.

It's an interesting question which ancient populations contributed to the European Jewish gene pool. I had a go at figuring this out a few times with the currently available ancient data, and basically gave up, because it's just too complicated. I'm looking forward to those Philistine samples.

kingjohn
06-10-2018, 08:02 AM
You guys should use the Global25 for this, because it picks up recent drift better than the BR K7.

It's an interesting question which ancient populations contributed to the European Jewish gene pool. I had a go at figuring this out a few times with the currently available ancient data, and basically gave up, because it's just too complicated. I'm looking forward to those Philistine samples.

davdiski what do you see in your global k25 runs for aschenazi and sefharadi ?

Generalissimo
06-10-2018, 08:08 AM
davdiski what do you see in your global k25 runs for aschenazi and sefharadi ?

I see a lot of things, but I don't know which of them are correct. :P

We just have to wait for the relevant ancient data.

Sikeliot
06-10-2018, 11:02 AM
Ashkenazim have significant Levantine input, the rest mostly derive their "Levantine" ancestry from Bronze age Anatolian populations. Compared to Bronze Age Sicilian samples modern Sicilians have + 10-15% North African and Levantine which is rather low and the extra Steppe balance out the Near Eastern so in other words modern Sicilians became both more Continental European and Near Eastern still don't really differ from Bronze age samples - Mainland Greeks on the other hand received significant Steppe or let's say some Slavic input that changed their position compared to Bronze age Hellenic samples.

It is difficult to measure the true genetic position of the Bronze Age Sicilian because it had a relatively low SNP call rate, and I know it was disputed by Erikl86 but on many calculators the Bronze Age Sicilian result displayed a closeness with North African Jews and Sephardim which implied to me that a significant portion of the Near Eastern DNA was already present well before Phoenicians landed on their shores. I also maintained that since the Bell Beaker sample came from a region under influence from Sardinia (which is where Bell Beaker spread to Sicily from), you could likely find people in other parts of the island in that era who were even more eastern Mediterranean.

MDLP World:

MDLP World:
# Population Percent
1 South_and_West_European 45.36
2 Middle_East 35.67
3 Caucaus_Parsia 13.16
4 North_and_East_European 4.13
5 North_Asian 1.33


Finished reading population data. 257 populations found.
12 components mode.

--------------------------------

Least-squares method.

Using 1 population approximation:
1 Jew_Morocco @ 7.166730
2 Jew_France @ 8.249454
3 Sephardim @ 8.596121
4 Jew_Italia @ 8.938586
5 Jew_Algeria @ 10.401005
6 Jew_Tunisia @ 11.487771
7 Jew_Libya @ 12.373126
8 Ashkenazim @ 13.519056
9 Sicilian @ 14.205534
10 Greek_Cretan @ 17.168140
11 Italian-South @ 17.168957
12 Jew_Romania @ 18.039038
13 Cypriot @ 18.323494
14 Italian-Center @ 18.755230
15 Greek_East @ 19.533710
16 Jew_Syria @ 20.221943
17 Greek_Center @ 20.958565
18 Greek_South @ 22.544491
19 Ashkenazim_V @ 22.830824
20 Greek_North @ 23.659155

Using 2 populations approximation:
1 50% Samaritian +50% Sardinian @ 3.602658



MDLP World22:

# Population Percent
1 Atlantic_Mediterranean_Neolithic 41.36
2 Near_East 26.70
3 West-Asian 22.43
4 North-East-European 4.58
5 North-European-Mesolithic 2.52
6 Austronesian 1.85


Finished reading population data. 276 populations found.
22 components mode.

--------------------------------

Least-squares method.

Using 1 population approximation:
1 Sephardim_derived @ 5.335933
2 Jew_Algeria_derived @ 6.473939
3 Jew_Italia_derived @ 6.937978
4 Jew_Francestrale_derived @ 7.165184
5 Jew_Morocco_derived @ 9.023011
6 Jew_Tunisia_derived @ 9.559836
7 Jew_Libya_derived @ 10.691164
8 Ashkenazim_derived @ 11.281357
9 Sicilian_derived @ 12.441468
10 Cypriot_derived @ 13.258138
11 Greek_Cretan_derived @ 13.432624
12 Jew_Syria_derived @ 14.666184
13 Italian-South_derived @ 15.173383
14 Italian-Center_derived @ 17.368025
15 Greek_East_derived @ 19.032942
16 Jew_Romania_derived @ 19.691177
17 Greek_Center_derived @ 20.506847
18 Lebanese_derived @ 22.316582
19 Costanoan_derived @ 22.689951
20 Druze_derived @ 23.163324

Using 2 populations approximation:
1 50% Jew_Azerbaijan_derived +50% Otzi_derived @ 3.807522

Using 3 populations approximation:
1 50% Jew-Iraqi_derived +25% Otzi_derived +25% Sardinian_derived @ 3.038447



Dodecad K7b:


Dodecad K7b:
# Population Percent
1 Southern 46.58
2 Atlantic_Baltic 33.93
3 West_Asian 18.13


Finished reading population data. 223 populations found.
7 components mode.

--------------------------------

Least-squares method.

Using 1 population approximation:
1 Morocco_Jews_Behar @ 10.226101
2 Ashkenazi_Dodecad @ 10.331970
3 Ashkenazy_Jews_Behar @ 10.906626
4 Sicilian_Dodecad @ 11.025823
5 S_Italian_Sicilian_Dodecad @ 11.577702
6 Sephardic_Jews_Behar @ 12.401764
7 C_Italian_Dodecad @ 15.393037
8 Greek_Dodecad @ 16.089119
9 TSI30_Metspalu @ 20.511086
10 Tuscan_HGDP @ 20.860069
11 O_Italian_Dodecad @ 23.185001
12 Cypriots_Behar @ 23.607611
13 Algerian_Dodecad @ 24.849754
14 Sardinian_HGDP @ 26.124981
15 Moroccan_Dodecad @ 26.234802
16 North_Italian_HGDP @ 27.556782
17 Canarias_1000Genomes @ 28.298319
18 N_Italian_Dodecad @ 28.830986
19 Lebanese_Behar @ 30.180227
20 Bulgarian_Dodecad @ 31.055723

HarappaWorld:

HarappaWorld:
# Population Percent
1 Mediterranean 39.12
2 Caucasian 36.00
3 SW-Asian 12.12
4 NE-Euro 7.54
5 Baloch 3.10
6 Pygmy 2.12


Finished reading population data. 377 populations found.
16 components mode.

--------------------------------

Least-squares method.

Using 1 population approximation:
1 morocco-jew_behar @ 15.113560
2 sephardic-jew_behar @ 15.609380
3 ashkenazy-jew_behar @ 17.855648
4 ashkenazi_harappa @ 19.632252
5 tuscan_1000genomes @ 19.884975
6 tuscan_hgdp @ 20.145630
7 tuscan_hapmap @ 20.264194
8 cypriot_behar @ 24.902407
9 italian_hgdp @ 26.690432
10 sardinian_hgdp @ 27.687870
11 turk-aydin_hodoglugil @ 29.055609
12 lebanese_behar @ 30.591202
13 turk-kayseri_hodoglugil @ 32.443691
14 lebanese-muslim_haber @ 32.940960
15 turk-istanbul_hodoglugil @ 33.070450
16 jordanian_behar @ 33.554523
17 lebanese-christian_haber @ 33.989632
18 syrian_behar @ 34.128044
19 turk_behar @ 34.129539
20 bulgarian_yunusbayev @ 34.185703



Ancient Eurasia K6:

Ancient Eurasia K6:
# Population Percent
1 Natufian 50.54
2 West_European_Hunter_Gartherer 23.58
3 Ancestral_South_Eurasian 11.49
4 Ancestral_North_Eurasian 8.37
5 Sub_Saharan 6.02


Using 1 population approximation:
1 Jew_Libyan @ 13.853441
2 Jew_Tunisian @ 14.513676
3 Jew_Ashkenazi @ 15.022666
4 Jew_Moroccan @ 15.138251
5 Sicilian @ 16.429512
6 Italian_South @ 17.431561
7 Greek @ 18.535622
8 Albanian @ 19.021828
9 Libyan @ 19.204952
10 Cypriot @ 19.380522
11 Jordanian @ 19.384632
12 BedouinA @ 19.415173
13 Palestinian @ 19.510408
14 Egyptian @ 19.527651
15 Lebanese @ 19.564959
16 Saudi @ 20.471405
17 Syrian @ 20.594830
18 Armenia_ChL @ 20.724758
19 Levant_BA @ 20.959898
20 Druze @ 21.221704


Additionally, there are some Sicilians today who almost entirely lack the Steppe component and may not differ much from their ancient ancestors. If you look at this chart from Sarno et al, which is not averages but individual bars, you can see a few people from Palermo (SSI_SIC_PA), Catania (SSI_SIC_CT), Ragusa (SSI_SIC_RG) and Reggio Calabria (SSI_CAL_RC) who do not have any of the blue Steppe/"European" component, for whom the red bar (Near Eastern) drops all the way to the bottom. Blue is Steppe, green is Sardinian/EEF, red is Levant/Arabian and yellow is Caucasus/CHG.This is clearly not the case for any of the mainland Greeks or Albanians toward the left of the chart. Now, I do believe there is recent MENA input into Sicily, mostly in the west of the island -- the North African component and some instances of elevated SW Asian components clearly did not arrive from Anatolia -- but I maintain some of it was present in the Bronze Age sample.

My opinion is that the aforementioned regions of southern Italy have either:

1. absorbed some degree of extra Near Eastern compared to Bronze Age Sicilian samples we have now,
2. descend from a less Sardinian-admixed Bronze Age population on the island that we have not yet ancient samples for,

and remained fairly locked in time, having not absorbed any populations that were high in Steppe. The non-native components in south Italy that can most clearly signal foreign admixture are North European and North African.

https://i.imgur.com/2pJaHeR.png



That is true in the case of some genetic outlier samples from Northern Greece. Peloponnese and Southern Greeks would be more Myceanean and less slavic than that. Islanders and Western Asia Minor Greeks have higher Myceanean than any Mainlander which answer some of your previous questions.

I am of the opinion that all mainland Greeks, with the possible exception of some Tsakonians and Maniots, have some degree of Slavic input which is highest in Macedonia, Thessaly, Epirus, and Thrace and lower as you go south from that. But, Islanders and Asia Minor Greeks however did absorb extra West Asian admixture. It is not JUST that the mainlanders absorbed Slavic. Both groups diverged in opposite directions.


Yes Minoans were related to other Neolithic-derived Europeans, including pre-Greek colonization BA Sicily. I would say the closest living Europeans genetically would be modern-day Sardinians.

Minoans, like the Bronze Age Sicilian sample, seem to be best modeled as varying mixtures of Sardinian and West Asian. I would like to see other Bronze Age Sicilian samples, from the central and eastern parts of the island that did not see Bell Beaker influence, there is no reason to believe they would differ from Minoans.

This is also why I think some of what is being modeled as Mycenaean in Sicily is actually their native ancestry.


You guys should use the Global25 for this, because it picks up recent drift better than the BR K7.

It's an interesting question which ancient populations contributed to the European Jewish gene pool. I had a go at figuring this out a few times with the currently available ancient data, and basically gave up, because it's just too complicated. I'm looking forward to those Philistine samples.

Philistines should come out similarly to Minoans, Bell Beaker Sicilian, and so on.


Unlikely, as Hurrians would have brought this earlier. Haber et al. himself suggests this happened 2900 +- ~750 years ago, after the Mitanni were already gone. And I don't know if Hurrians and Mitanni actually had significant Steppes-like admixture.

I still don't understand why you reject the possibility of substantial Greek admixture among Levantines. They have been settling in the Levant since the 10th century BC and Greek became lingua franca at one point in that region along side Aramaic. Also, Judaism itself was very popular among 1st century Greco-Romans.

Cultural, not genetic shift. If the Levant did not become genetically Saudi Arabian after the Arab conquest, which was far more culturally influential, why would they have become transplanted Greeks? Keep in mind also that the Jews resisted Hellenization and would not have been likely to admix with them substantially. One of the Jewish holidays is even based on the premise of their resisting Hellenization. ;)

Kurd
06-10-2018, 11:40 AM
The confounding issue with determining steppe ancestry in populations with substantial W Asian ancestry is that EMBA and MLBA groups have substantial W Asian ancestry themselves. Thus results from allele sharing programs such as ADMIXTURE, as well as OUTGROUP mode f3s and f4s can't be relied upon to determine EMBA or MLBA steppe genetic input in those groups.

I have developed a very robust pipeline using multiple lines of evidence using formal stats that circumvents the above issues, and I was able to use it to determine the type and amount of steppe ancestry in Kurds. Can't leak more on the outcome here, because it was a 2 week project, and the results will be published in a comprehensive article.

In general terms this is how the tests should be conducted:

1- Determine 2 or more populations perhaps Chalcolithic (say J) that we are confident that form the genetic substructural backbone of jews;

2- Use dstats of the form D [ J, modern jew; steppe, Mbuti ] for many EMBA, MLBA, IA, Middle-Age steppe pops, to see if modern jews are any more steppe shifted than those 2 or 3 chalcolithic pops (J), and which groups cause the most shift.

If the results indicate modern jews are not for any of those pops ( which I seriously doubt) then jew don't have steppe admixture.

3- Confirm the above findings with f3s of the form f3 [ J, steppe; modern jews ] and note which pairs produce the maximum signal of admixture

4- Take the best candidates and run qpAdm checking that you have a reasonable p-value and reasonable standard errors.

NOTE: Since many Chalcolithic source pops (J) are quite related to each other, you would probably have to introduce certain Chalcolithic outgroups into qpAdm that allow you to adequately parse the sources.


EDIT: the above steps can be modified for Aegean and Levantine admixture

Erikl86
06-10-2018, 12:12 PM
davdiski what do you see in your global k25 runs for aschenazi and sefharadi ?

I see the following, extremely interesting things:

First of all, the closest I could got modelling Samaritan is this:

[1] "distance%=1.4113 / distance=0.014113"

Samaritan

Levant_BA:I1730 49.5
Assyrian:Assyrian165 21.1
Assyrian:Assyrian151 18.6
Assyrian:Assyrian155 10.8

Compare to just Levant_BA:

[1] "distance%=3.0079 / distance=0.030079"

Samaritan

Levant_BA:I1730 100
Levant_BA:I1705 0
Levant_BA:I1706 0


Also, Samaritans gave bad results when tried to model with Anatolian, which they should have also if it came from Hittites, Hurrians or Mitanni - since they pre-date the formation of the Samaritans themselves as a separate people from the rest of the Canaanites:

[1] "distance%=1.3833 / distance=0.013833"

Samaritan

Levant_BA:I1730 49.10
Assyrian:Assyrian165 19.60
Assyrian:Assyrian151 18.10
Assyrian:Assyrian155 8.45
Anatolia_IA:MA2197 4.75
Levant_BA:I1705 0.00
Levant_BA:I1706 0.00
Assyrian:Assyrian152 0.00
Assyrian:Assyrian153 0.00
Assyrian:Assyrian159 0.00
Assyrian:Assyrian160 0.00
Assyrian:Assyrian161 0.00
Assyrian:Assyrian162 0.00
Assyrian:Assyrian163 0.00
Assyrian:Assyrian164 0.00
Anatolia_BA:I2495 0.00
Anatolia_BA:I2683 0.00
Anatolia_IA:MA2198 0.00


Also, trying to model Samaritans with just Mycenaeans and no Assyrians gave a bogus result:

[1] "distance%=2.0235 / distance=0.020235"

Samaritan

Levant_BA:I1730 63.30
Mycenaean:I9033 31.15
Levant_BA:I1706 5.55


Using Levant_BA gives +3% distance, x3 times as large. I couldn't model the qpAdm suggested by Hebar et al. and Lazirdis because there is no CHL_Iran in the populations. I guess I could replace it with something else, but that won't be the same exact model, but I have used Hajji_Firuz CHL stats because it's a CHL Iranian sample, which appear to give better distances than Levant_BA:

[1] "distance%=1.2174 / distance=0.012174"

Samaritan

Levant_N:I1699 42.0
Hajji_Firuz_ChL:I2323 31.4
Hajji_Firuz_ChL_o:I2327 24.1
Hajji_Firuz_ChL:I4351 2.6

Again, trying to incorporate BA and IA Anatolians (Hitties, Hurrians, Mitanni):

[1] "distance%=1.2174 / distance=0.012174"

Samaritan

Levant_N:I1699 42.0
Hajji_Firuz_ChL:I2323 31.4
Hajji_Firuz_ChL_o:I2327 24.1
Hajji_Firuz_ChL:I4351 2.6
Hajji_Firuz_ChL:I4241 0.0
Hajji_Firuz_ChL:I4349 0.0
Levant_N:I0867 0.0
Anatolia_BA:I2495 0.0
Anatolia_BA:I2683 0.0
Anatolia_IA:MA2197 0.0
Anatolia_IA:MA2198 0.0

Didn't change much.


This also gave me better results when I used Assyrians:

[1] "distance%=1.1446 / distance=0.011446"

Samaritan

Levant_N:I1699 37.3
Hajji_Firuz_ChL:I2323 16.2
Hajji_Firuz_ChL_o:I2327 12.4
Assyrian:Assyrian151 11.9
Assyrian:Assyrian155 11.4
Assyrian:Assyrian165 10.8


And seems more realistic - ~35% Assyrian / 65% BA Levantine, because Samaritans cluster closer to BA Levantines than to Assyrians.

Also, when I use Levant BA instead of this qpAdm model, Mycenaeans and Minoans appear (which contradicts what I got in the K7 BR calculator), but when I use this qpAdm model of Levant_N/CHL_Iran, it doesn't:

[1] "distance%=1.1446 / distance=0.011446"

Samaritan

Levant_N:I1699 37.3
Hajji_Firuz_ChL:I2323 16.2
Hajji_Firuz_ChL_o:I2327 12.4
Assyrian:Assyrian151 11.9
Assyrian:Assyrian155 11.4
Assyrian:Assyrian165 10.8
Hajji_Firuz_ChL:I4241 0.0
Hajji_Firuz_ChL:I4349 0.0
Hajji_Firuz_ChL:I4351 0.0
Levant_N:I0867 0.0
Assyrian:Assyrian152 0.0
Assyrian:Assyrian153 0.0
Assyrian:Assyrian159 0.0
Assyrian:Assyrian160 0.0
Assyrian:Assyrian161 0.0
Assyrian:Assyrian162 0.0
Assyrian:Assyrian163 0.0
Assyrian:Assyrian164 0.0
Mycenaean:I9006 0.0
Mycenaean:I9010 0.0
Mycenaean:I9033 0.0
Mycenaean:I9041 0.0

As opposed to using Levant_BA:


[1] "distance%=1.3269 / distance=0.013269"

Samaritan

Levant_BA:I1730 47.45
Assyrian:Assyrian151 16.85
Assyrian:Assyrian165 15.85
Mycenaean:I9033 9.60
Assyrian:Assyrian155 8.45
Levant_BA:I1706 1.80

And also with Minoans:

[1] "distance%=1.3381 / distance=0.013381"

Samaritan

Levant_BA:I1730 42.15
Assyrian:Assyrian151 17.75
Assyrian:Assyrian155 15.35
Assyrian:Assyrian165 13.30
Minoan_Lasithi:I0074 7.70
Levant_BA:I1706 3.75

So, if Levant_BA is better than my new improvised N_Levant/Hajji_Firuz_ChL model, it seems Samaritans do have a small amount of Aegean admixture, either because they did take some converts early on during Hellenistic times, or from Philistines via Judeans.

Next, tried to model Lebanese Christians. The closest distance I got using Levant_BA, was 1.4% with Mycenaean and Minoans:

[1] "distance%=1.6354 / distance=0.016354"

Lebanese_Christian

Levant_BA:I1705 42.1
Mycenaean:I9033 41.5
Levant_BA:I1730 14.8
Minoan_Lasithi:I0074 1.5

Not much different than if I use only Mycenaean:

[1] "distance%=1.6365 / distance=0.016365"

Lebanese_Christian

Mycenaean:I9033 42.8
Levant_BA:I1705 41.6
Levant_BA:I1730 15.6

When I tried to put Anatolian instead of Mycenaean, I got a little better distance, but the results are complete bogus:

[1] "distance%=1.0896 / distance=0.010896"

Lebanese_Christian

Anatolia_BA:I2683 30.6
Anatolia_IA:MA2198 24.8
Levant_BA:I1730 19.9
Levant_BA:I1706 16.8
Anatolia_BA:I2495 8.0

Only 35% Levantine is contrary to current findings. Plus, the fact that Samaritans, which should have also been mixed with Anatolians, got insignificant admixture of Anatolians, kind of discard that it came from BA Anatolian and not directly from Mycenaean.

Now, when I try to use the Levant_N/Hajji_Firuz_ChL model, I get much closer results:

[1] "distance%=0.7406 / distance=0.007406"

Lebanese_Christian

Hajji_Firuz_ChL_o:I2327 31.1
Levant_N:I1699 30.4
Mycenaean:I9041 13.8
Hajji_Firuz_ChL:I4241 11.6
Hajji_Firuz_ChL:I4351 7.0
Hajji_Firuz_ChL:I2323 6.1

As opposed to:

[1] "distance%=0.7953 / distance=0.007953"

Lebanese_Christian

Levant_N:I1699 30.70
Hajji_Firuz_ChL_o:I2327 28.00
Anatolia_BA:I2683 15.70
Hajji_Firuz_ChL:I4241 11.55
Anatolia_IA:MA2198 6.70
Hajji_Firuz_ChL:I4351 6.00
Hajji_Firuz_ChL:I2323 1.35

Again, the fact that Samaritans, which should have also been mixed with Anatolians, got almost no admixture of Anatolians when using the Levant_N/CHL_Iran and Levant_BA, kind of discard that it came from BA Anatolian and not directly from Mycenaean. I find it very difficult to believe that for some weird reason it got up from nothing to 20% for those specific Canaanite groups, when after the 10th century BC, there were no significant Anatolian-centered empires in the Levant.

So, it seems by using Global25, Aegean admixture among Lebanese Christians is ~14% or ~42%. Now, let's see how much Steppes-like admixture does Global25 predict for Mycenaeans:

Sintashta as Steppes-like admixture:

[1] "distance%=1.0953 / distance=0.010953"

Mycenaean

Minoan_Lasithi:I0070 73.30
Minoan_Lasithi:I0074 9.55
Sintashta_MLBA:I1012 9.00
Sintashta_MLBA:I1022 8.15

vs. Srubnaya as Steppes-like admixture:

[1] "distance%=1.1378 / distance=0.011378"

Mycenaean

Minoan_Lasithi:I0070 72.30
Minoan_Lasithi:I0074 10.00
Srubnaya_MLBA:I0359 9.65
Srubnaya_MLBA:I0430 6.90
Srubnaya_MLBA:I0424 1.15


So, again, just like in K7-BR data, I got 80/20 for Minoan and Steppes like.

Now, I'm suspicious here, because, according to Hebar et al., there should be ~7% Steppes-like admixture in modern Lebanese that didn't exist in BA Levantines. If I use Levant BA, then 20% from the ~42% Mycenean I got is ~8%.

But 20% from 14% is just too low, and doesn't fit what Haber et al. found. So despite the short distance, this is a first sign it might not be the best model, and maybe just a case of over-fit.

Now, let's try Western Jews:

First, Sephardic Jew:


[1] "distance%=0.5648 / distance=0.005648"

Sephardic_Jew

Hajji_Firuz_ChL_o:I2327 23.65
French_South:SouthFrench3326 13.70
Mycenaean:I9033 12.90
Levant_N:I1699 12.10
Hajji_Firuz_ChL:I4351 11.40
Minoan_Lasithi:I0073 9.85
Moroccan:MCA14 7.70
Moroccan:MCA7 5.25
French_South:SouthFrench3951 3.45


vs. Italian:

[1] "distance%=0.6937 / distance=0.006937"

Sephardic_Jew

Italian_Tuscan:NA20506 28.75
Hajji_Firuz_ChL:I2323 18.55
Hajji_Firuz_ChL:I4351 14.60
Levant_N:I1699 9.80
French_South:SouthFrench3326 9.20
Moroccan:MCA7 7.35
Moroccan:MCA14 6.10
Italian_Tuscan:NA20505 4.60
Moroccan:MCA9 1.05

When I use BA Levant instead of the qpAdm model, this is what I get:

[1] "distance%=1.3424 / distance=0.013424"

Sephardic_Jew

Mycenaean:I9033 45.85
Levant_BA:I1705 21.15
French_South:SouthFrench1323 13.05
Moroccan:MCA7 10.55
Levant_BA:I1730 7.65
Moroccan:MCA14 1.75

And with Minoan:


[1] "distance%=1.3366 / distance=0.013366"

Sephardic_Jew

Mycenaean:I9033 42.75
Levant_BA:I1705 21.95
French_South:SouthFrench1323 13.40
Moroccan:MCA7 10.95
Levant_BA:I1730 6.10
Minoan_Lasithi:I0074 3.15
Moroccan:MCA14 1.70


Moroccan and French stays the same, but Mycenaean vs Levantine is much closer than to what I got on K7-BR. The distance though is again worse.

vs. Italian:

[1] "distance%=1.4358 / distance=0.014358"

Sephardic_Jew

Italian_Tuscan:NA20508 39.10
Levant_BA:I1730 18.35
Italian_Tuscan:NA20506 12.00
Moroccan:MCA14 9.85
Italian_Tuscan:NA20502 9.05
Italian_Tuscan:NA20505 7.45
Moroccan:MCA7 4.20

Which is bogus results, because then Sephardic Jews have only 18% Levantine ancestry.


And when using only modern populations, and employing Greek Cypriots instead of the Levant_N/CHL_Iran model, I got much better results than both:

[1] "distance%=0.5541 / distance=0.005541"

Sephardic_Jew

Cypriot:CYP2 32.25
Cypriot:Cyprus24AJ19 22.50
Cypriot:CYP5 13.30
French_South:SouthFrench3326 12.15
Moroccan:MCA7 8.10
French_South:SouthFrench1112 5.05
Cypriot:Cyprus21AJ19 4.00
Moroccan:MCA14 2.65

Which made me test how Cypriots can be best modeled in Global25 - after reading the Lazirdis article, we know that it should have about ~20-30% Levantine admixture. I've tried to model it both with this improvised qpAdm model, and with Levant_BA, to know which one is better, being that again I don't have CHL_Iranian in Global25 and had to use Hajji_Firuz_ChL instead:

Cypriot with Levant_BA:


[1] "distance%=1.3817 / distance=0.013817"

Cypriot

Mycenaean:I9033 53.00
Levant_BA:I1705 37.75
Levant_BA:I1730 7.05
Minoan_Lasithi:I0074 2.20

So, it's 45% Levantine, 55% Aegean.

Cypriot with Levant_N/Hajji_Firuz_ChL:


[1] "distance%=0.5397 / distance=0.005397"

Cypriot

Hajji_Firuz_ChL_o:I2327 24.20
Mycenaean:I9041 21.95
Levant_N:I1699 18.90
Hajji_Firuz_ChL:I4351 18.30
Mycenaean:I9033 10.20
Hajji_Firuz_ChL:I4241 4.15
Levant_N:I0867 2.30

So, it seems this model, despite the fact that it yields some good results, is worst than BA Levant, because obviously Cypriots are not only 32% Aegean. Too bad we don't have the CHL_Iran model :-( . But, I'll take the next best thing - this Levant_BA.

Now, in Global25, we have Italian Jews. Let's try to model them with Mycenaeans and Minoans:

[1] "distance%=1.4467 / distance=0.014467"

Italian_Jew

Mycenaean:I9033 59.6
Levant_BA:I1705 22.5
Levant_BA:I1706 15.8
Minoan_Lasithi:I0074 2.1

And now only with Mycenaeans:

[1] "distance%=1.4495 / distance=0.014495"

Italian_Jew

Mycenaean:I9033 61.5
Levant_BA:I1705 23.1
Levant_BA:I1706 15.3

Basically, for both of them, it's 40/60 Levant_BA / ancient Aegeans.

Next, with Tuscans:

[1] "distance%=1.2185 / distance=0.012185"

Italian_Jew

Italian_Tuscan:NA20508 44.0
Levant_BA:I1730 29.5
Italian_Tuscan:NA20505 26.5


That is even more preposterous as there is no chance Italkim are just 30% Levantines and 70 Tuscans. Even if the distance is better.

And now, instead of Levant BA, they should be pretty close to Cypriots, as basically they should lack the French or Moroccan admixture seen in Sephardic Jews:

[1] "distance%=1.1847 / distance=0.011847"

Italian_Jew

Cypriot:Cyprus24AJ19 35.10
Cypriot:CYP2 20.40
Cypriot:Cyprus13AJ19 19.75
Cypriot:Cyprus22AJ19 19.20
Cypriot:CYP5 3.25
Cypriot:Cyprus21AJ19 2.30

Better results, but this gave me idea. I want to get the same distance I scored when I modelled Sephardic Jews with Cypriots. Perhaps, there is some Tuscan admixture, the result of Italkim living for centuries in Italy, while Sephardim and Ashkenazim moved north to France.

[1] "distance%=0.8507 / distance=0.008507"

Italian_Jew

Italian_Tuscan:NA20508 29.8
Cypriot:Cyprus24AJ19 26.0
Cypriot:Cyprus22AJ19 16.2
Cypriot:CYP2 15.2
Cypriot:CYP5 7.6
Cypriot:Cyprus2AJ19 3.3
Italian_Tuscan:NA20504 1.9

Shorter distance, and again, Western Jews are about 50% Aegean, 30% Levantines, consider that Tuscans were modeled by Davidsky as:


Italian_Tuscan

England_Roman_outlier 0.118±0.115
Mycenaean 0.521±0.147
Unetice 0.361±0.059

P-value 0.741956816
chisq 9.402

That is, about 50% correlation with Mycenaeans, and about 10% Levantine (Roman outlier in England is Levantine).

Now, for Ashkenazi Jews. Because I have Italkim samples, I'd like to see how close they are. Best one I got was with modern Italian Jews + North/East Europeans. Single nMonte from Italian Jews was very similar to modern Cypriots:

[1] "distance%=0.3019 / distance=0.003019"

Ashkenazi_Jew

Italian_Jew:ItalyJew8 25.00
Italian_Jew:ItalyJew6 16.00
Italian_Jew:ItalyJew1 14.05
Italian_Jew:ItalyJew4 12.85
Italian_Jew:ItalyJew7 8.55
Italian_Jew:ItalyJew5 8.15
Polish:Poland6 5.35
Italian_Jew:ItalyJew10 4.55
German:Germany24 2.25
German:Germany22 2.10
Italian_Jew:ItalyJew3 1.15


Got similar results when used Cypriots + North/East Europeans, ones which make even more sense, because North/Eastern European admixture is more reasonable.

I suspect the northern admixture in Tuscans that can be found in Italkim casuses overlap and minimizes the admixture for Ashkenazi Jews:

[1] "distance%=0.87 / distance=0.0087"

Ashkenazi_Jew

Cypriot:Cyprus24AJ19 28.05
Cypriot:Cyprus22AJ19 24.20
Polish:Polish203 10.20
Lithuanian:LithuanianD1 9.15
Cypriot:CYP19 8.65
Cypriot:Cyprus21AJ19 7.05
Cypriot:CYP2 6.45
Cypriot:Cyprus2AJ19 6.25

Or basically 80% Cypriot, 20% North/Eastern European. Which would be equal to 35% Levantine, 45% Aegean and 20% North/East European.

When modeled using Levant BA + ancient Aegeans instead of Cypriots, + modern North/East Europeans:

[1] "distance%=1.1992 / distance=0.011992"

Ashkenazi_Jew

Mycenaean:I9033 39.05
Levant_BA:I1730 21.25
Polish:Polish203 19.05
Levant_BA:I1706 7.70
Levant_BA:I1705 6.50
Ukrainian:595_R01C02 6.45

When tries to model Levant_BA + Tuscans + North/East Europeans, got this:

[1] "distance%=1.1012 / distance=0.011012"

Ashkenazi_Jew

Italian_Tuscan:NA20508 51.7
Levant_BA:I1730 26.9
Polish:Poland3 11.2
Italian_Tuscan:NA20505 10.2

This is bogus result, as it appears to suggest that Ashkenazi Jews are +60% Tuscans. Again, I think overlap of Germanic/Celtic with modern Tuscans, as well as a bit of Levantine and Aegean, screws up the results.



All in all, I must say that using Global25 gives a lot more North/East European ancestry when using ancient populations as base for Levantine + S. European admixture. In anyway, in all Jewish models, using ancient Aegeans or Cypriots as a base model yielded much more logical admixtures than using Tuscans.


So, to summarize it all, it seems that although the distances are bigger, the percentages are a little bit different, it's pretty consistent with what I got in K7-BR. The main surprise I got here is:

Cypriots seem to be much more Levantines.

Samaritans have a very small (less than 10%) ancient Aegean admixture.

Lebanese Christians' Steppes-like ancestry makes sense being that they get ~40% Myceneaen.

Italkim seem to be amixed with Italians + aegeans - make sense.

Sephardic Jews seem to be much more admixed with N. Africans - I guess it depends on which Sephardic Jews were sampled for the average.

Ashkenazi Jews seem much more admixed with North/Eastern European using Global25 (unless I use Cyprtios as a base model).

Sikeliot
06-10-2018, 12:36 PM
What if what is being proxied as Mycenaean/Minoan actually came from an Italian population from ancient times that we do not yet have ancient samples for?

kingjohn
06-10-2018, 12:53 PM
The confounding issue with determining steppe ancestry in populations with substantial W Asian ancestry is that EMBA and MLBA groups have substantial W Asian ancestry themselves. Thus results from allele sharing programs such as ADMIXTURE, as well as OUTGROUP mode f3s and f4s can't be relied upon to determine EMBA or MLBA steppe genetic input in those groups.

I have developed a very robust pipeline using multiple lines of evidence using formal stats that circumvents the above issues, and I was able to use it to determine the type and amount of steppe ancestry in Kurds. Can't leak more on the outcome here, because it was a 2 week project, and the results will be published in a comprehensive article.

In general terms this is how the tests should be conducted:

1- Determine 2 or more populations perhaps Chalcolithic (say J) that we are confident that form the genetic substructural backbone of jews;

2- Use dstats of the form D [ J, modern jew; steppe, Mbuti ] for many EMBA, MLBA, IA, Middle-Age steppe pops, to see if modern jews are any more steppe shifted than those 2 or 3 chalcolithic pops (J), and which groups cause the most shift.

If the results indicate modern jews are not for any of those pops ( which I seriously doubt) then jew don't have steppe admixture.

3- Confirm the above findings with f3s of the form f3 [ J, steppe; modern jews ] and note which pairs produce the maximum signal of admixture

4- Take the best candidates and run qpAdm checking that you have a reasonable p-value and reasonable standard errors.

NOTE: Since many Chalcolithic source pops (J) are quite related to each other, you would probably have to introduce certain Chalcolithic outgroups into qpAdm that allow you to adequately parse the sources.


EDIT: the above steps can be modified for Aegean and Levantine admixture

people should follow your k53 calculator the last word in the field at the moment

Erikl86
06-10-2018, 01:08 PM
What if what is being proxied as Mycenaean/Minoan actually came from an Italian population from ancient times that we do not yet have ancient samples for?

In order to to prove/disprove this and other theories regarding Ashkenazi Jews' origins (and Western Jews in general), the following is a must:

1. Samples of 1st century Judean Jews.

2. Samples of 1st century Romans from Umbria or Tuscany.

3. Samples of 2nd century (that is, after the destruction of the temple) Roman Jews.

4. Sample of Philistines.

Philistines should be available soon, I'm unsure why we don't have non-Jewish Roman samples yet.

kingjohn
06-10-2018, 01:14 PM
In order to to prove/disprove this and other theories regarding Ashkenazi Jews' origins (and Western Jews in general), the following is a must:

1. Samples of 1st century Judean Jews.

2. Samples of 1st century Romans from Umbria or Tuscany.

3. Samples of 2nd century (that is, after the destruction of the temple) Roman Jews.

4. Sample of Philistines.

Philistines should be available soon, I'm unsure why we don't have non-Jewish Roman samples yet.

no rush for 1 now we got Canaanite lebanese genomes this more or less should be like the ancient judeans

Erikl86
06-10-2018, 01:18 PM
no rush for 1 now we got Canaanite lebanese genomes this more or less should be like the ancient judeans

Not necessarily, since Hebar et al. found that there seem to be a difference between modern day Levantines and those ancient Canaanites because of additional admixture, which he dates that happened somewhere between 1000 BC to the 1st century.

If we can examine 1st century Judean samples, we might know if Judean Jews indeed intermarried with Aegeans or Romans or how non-Levantine was the Levant back then.

kingjohn
06-10-2018, 01:35 PM
Not necessarily, since Hebar et al. found that there seem to be a difference between modern day Levantines and those ancient Canaanites because of additional admixture, which he dates that happened somewhere between 1000 BC to the 1st century.

If we can examine 1st century Judean samples, we might know if Judean Jews indeed intermarried with Aegeans or Romans or how non-Levantine was the Levant back then.

ok
can you run me i gave you my gedmatch kit
want to see how levantine+ aegean i am ?
use k36 it is also ok :)
kind regards
adam

Sikeliot
06-10-2018, 01:53 PM
Philistines should be available soon, I'm unsure why we don't have non-Jewish Roman samples yet.

Which modern peoples do you think they will be closest to?

Erikl86
06-10-2018, 01:54 PM
Using Global25 again for Ashkenazi Jews:

[1] "distance%=0.9715 / distance=0.009715"

Ashkenazi_Jew

Samaritan:GSM537033 36.6
Greek_Crete: B_Crete-2 33.3
Greek_Crete: NA17376 17.8
Polish: Polish203 12.3

vs.

[1] "distance%=0.5361 / distance=0.005361"

Ashkenazi_Jew

Italian_Tuscan:NA20508 26.75
Samaritan:168723 21.80
Samaritan:GSM537033 21.20
Italian_Tuscan:NA20504 8.95
Italian_Tuscan:NA20505 8.95
Italian_Tuscan:NA20506 5.85
Polish: Polish205 5.05
Polish: Poland1 1.45

^^^ bogus results (Ashkenazi Jews cannot be 6% North/East Europeans).

I've tried to model Cypriots with modern populations, this is what I got:

[1] "distance%=0.754 / distance=0.00754"

Cypriot

Greek_Crete: B_Crete-2 33.4
Samaritan: GSM537032 18.9
Samaritan: GSM537033 17.7
Greek_Crete: NA17376 17.1
Greek_Crete: B_Crete-1 13.0

Roughly 35% Samaritan vs. 65% Cretan Greeks.

Being that we know according to Lazirdis et al. 2017 that Cretans are roughly 10% Levantine-like, it should be translated as 40% Levantine, 60% Greek (Aegean), similar to what Global25 got with Levant_BA / Mycenaean admixture.

[1] "distance%=0.8928 / distance=0.008928"

Ashkenazi_Jew

Cypriot: Cyprus24AJ19 33.9
Cypriot: Cyprus22AJ19 28.4
Lithuanian: LithuanianD1 12.3
Cypriot: Cyprus21AJ19 12.0
Cypriot:Cyprus2AJ19 6.1
Polish: Polish203 5.9
Polish: Polish204 1.4

Cypriot = ~80% Cypriot, ~20% North/Eastern European.

80% Cypriot = 32% Levantine, 48% Aegean.

Sikeliot
06-10-2018, 01:59 PM
Using Global25 again for Ashkenazi Jews:

[1] "distance%=0.9715 / distance=0.009715"

Ashkenazi_Jew

Samaritan:GSM537033 36.6
Greek_Crete: B_Crete-2 33.3
Greek_Crete: NA17376 17.8
Polish: Polish203 12.3

vs.

[1] "distance%=0.5361 / distance=0.005361"

Ashkenazi_Jew

Italian_Tuscan:NA20508 26.75
Samaritan:168723 21.80
Samaritan:GSM537033 21.20
Italian_Tuscan:NA20504 8.95
Italian_Tuscan:NA20505 8.95
Italian_Tuscan:NA20506 5.85
Polish: Polish205 5.05
Polish: Poland1 1.45

^^^ bogus results (Ashkenazi Jews cannot be 6% North/East Europeans).

I've tried to model Cypriots with modern populations, this is what I got:

[1] "distance%=0.754 / distance=0.00754"

Cypriot

Greek_Crete: B_Crete-2 33.4
Samaritan: GSM537032 18.9
Samaritan: GSM537033 17.7
Greek_Crete: NA17376 17.1
Greek_Crete: B_Crete-1 13.0

Roughly 35% Samaritan vs. 65% Cretan Greeks.

Being that we know according to Lazirdis et al. 2017 that Cretans are roughly 10% Levantine-like, it should be translated as 40% Levantine, 60% Greek (Aegean), similar to what Global25 got with Levant_BA / Mycenaean admixture.

[1] "distance%=0.8928 / distance=0.008928"

Ashkenazi_Jew

Cypriot: Cyprus24AJ19 33.9
Cypriot: Cyprus22AJ19 28.4
Lithuanian: LithuanianD1 12.3
Cypriot: Cyprus21AJ19 12.0
Cypriot:Cyprus2AJ19 6.1
Polish: Polish203 5.9
Polish: Polish204 1.4

Cypriot = ~80% Cypriot, ~20% North/Eastern European.

80% Cypriot = 32% Levantine, 48% Aegean.


Try Sicily and mainland Greece for comparison if possible?

Kurd
06-10-2018, 03:21 PM
people should follow your k53 calculator the last word in the field at the moment

As of now its a K50. A couple of redundant pops were removed (LNBA, Scotland EBA, England N). Still since ADMIXTURE is ambigous to direction of geneflow, very carefully designed formal tests can flush out the truth, now that we have so many aDNA samples across time and space......

Erikl86
06-10-2018, 03:23 PM
Try Sicily and mainland Greece for comparison if possible?

I don't have modern Sicilians in Global25 for some reason, but I have S. Italians:

"distance%=1.2292 / distance=0.012292"

Ashkenazi_Jew

Italian_South:ITS7 38.4
Italian_South:BEL57 20.7
Italian_South:ITS2 19.2
Italian_South:ITS4 17.4
Lithuanian:LithuanianD1 4.3

^^^ bogus results.


Even though Ashkenazi Jew, on single mode, seems to cluster more with S. Italians than Cypriots, probably because S. Italians have more N. European admixture than Cypriots, which would better fit the admixture of Jews having 15-20% North/Eastern European admixture.

Plus, the distance is bigger.

Now let's try with Samaritans in the mix:

[1] "distance%=0.5801 / distance=0.005801"

Ashkenazi_Jew

Italian_South:ITS7 40.10
Samaritan:GSM537033 21.75
Samaritan:GSM537032 12.05
Italian_South:BEL57 10.90
Italian_South:ITS4 4.70
Italian_South:ITS2 3.65
Polish: Poland2 3.30
Polish: Polish203 2.25
Lithuanian:LithuanianD1 1.30

Again, bogus results - North/Eastern Europeans get only ~7%. Samaritans seem to get close to what we have when mixed with Cretans: ~34%. Remember, we know that Aegeans settled in S. Italy, not that Italians settled in the Aegean en mass.

Now, let's try to break apart S. Italians. I've tried to model it using S. Greeks and modern Aegeans, Samaritans, Tuscans and French (which would represent Normans):

[1] "distance%=0.5615 / distance=0.005615"

Italian_South

Italian_Tuscan:NA20505 35.30
Greek_Crete:B_Crete-2 19.05
Samaritan:168723 13.20
Samaritan:S_Samaritan-1 7.45
French:France1 6.10
French:France38 5.25
Italian_Tuscan:NA20502 3.50
Italian_Tuscan:NA20506 3.20
French:France35 3.05
Greek_Crete:B_Crete-1 2.45
Samaritan:GSM537033 1.45

Italian Tuscan: 42% (some of it Greek-derived, and Levantine-derived)
Cretan-like Greek: 21.5%
Levantine: 22%
French (aka Norman): ~14.5%

Now, let's try Italian Bergamo to cancel out any significant Levantine or Greek admixture which might exist in modern Tuscan Italians to some degree. Also, tried to add some Nordic populations to maybe emphasis Norman admixture:

[1] "distance%=0.6227 / distance=0.006227"

Italian_South

Greek_Crete:B_Crete-2 33.25
Italian_Bergamo:HGDP01151 15.95
Samaritan:168723 12.10
Italian_Bergamo:HGDP01153 9.20
Italian_Bergamo:HGDP01147 8.25
Greek_Crete:B_Crete-1 7.60
Greek_Crete:NA17376 4.40
French:France16 3.75
French:France9 3.55
Samaritan:GSM537033 1.95

Cretan-like Greek: 45.25% (should be translated to ~40%)
Levantine: 14% (should be translated to ~19%)
Northern Italian: 33.4% (how much is German/Celtic?)
French: 7.3%

Regarding the Greek - seems like I was correct - using Bergamo Italians, which are further north and are more likely less admixed with Greeks than Tuscans, elevated the amount of Greek admixture.

Which fit this model:

"Overall the estimated Balkan and North Western European paternal contributions in South Italy and Sicily are about 63% and 26% respectively."
https://en.wikipedia.org/wiki/Genetic_history_of_Italy#Y-DNA_introduced_by_historical_immigration

And, as I said - S. Italians are a mixture of Greeks, Northern Europeans and Levantines - which is the reason why they cluster so close to Western Jews. If Cypriots would have been more mixed with Northern Europeans, I believe Ashkenazi Jews would have clustered with them.


EDIT: Thought about modelling with Sardinians, as they represent older European genetic "form":

[1] "distance%=0.6307 / distance=0.006307"

Italian_South

Greek_Crete:B_Crete-2 43.90
French:France38 11.50
Sardinian:HGDP01073 10.15
French:France1 9.95
Greek_Crete:B_Crete-1 8.10
Samaritan:S_Samaritan-1 7.90
Samaritan:168723 5.70
Swedish:Sweden5 1.50
Greek_Crete:NA17376 1.30

Again, seem like Cretan-like Greek get's extremely high proportions.

Erikl86
06-10-2018, 03:25 PM
ok
can you run me i gave you my gedmatch kit
want to see how levantine+ aegean i am ?
use k36 it is also ok :)
kind regards
adam

I will try and send you in private.

Sikeliot
06-10-2018, 03:35 PM
I don't have modern Sicilians in Global25 for some reason, but I have S. Italians:

"distance%=1.2292 / distance=0.012292"

Ashkenazi_Jew

Italian_South:ITS7 38.4
Italian_South:BEL57 20.7
Italian_South:ITS2 19.2
Italian_South:ITS4 17.4
Lithuanian:LithuanianD1 4.3

^^^ bogus results.


Even though Ashkenazi Jew, on single mode, seems to cluster more with S. Italians than Cypriots, probably because S. Italians have more N. European admixture than Cypriots, which would better fit the admixture of Jews having 15-20% North/Eastern European admixture.

Plus, the distance is bigger.

Now let's try with Samaritans in the mix:

[1] "distance%=0.5801 / distance=0.005801"

Ashkenazi_Jew

Italian_South:ITS7 40.10
Samaritan:GSM537033 21.75
Samaritan:GSM537032 12.05
Italian_South:BEL57 10.90
Italian_South:ITS4 4.70
Italian_South:ITS2 3.65
Polish: Poland2 3.30
Polish: Polish203 2.25
Lithuanian:LithuanianD1 1.30

Again, bogus results - North/Eastern Europeans get only ~7%. Samaritans seem to get close to what we have when mixed with Cretans: ~34%. Remember, we know that Aegeans settled in S. Italy, not that Italians settled in the Aegean en mass.

Now, let's try to break apart S. Italians. I've tried to model it using S. Greeks and modern Aegeans, Samaritans, Tuscans and French (which would represent Normans):

[1] "distance%=0.5615 / distance=0.005615"

Italian_South

Italian_Tuscan:NA20505 35.30
Greek_Crete:B_Crete-2 19.05
Samaritan:168723 13.20
Samaritan:S_Samaritan-1 7.45
French:France1 6.10
French:France38 5.25
Italian_Tuscan:NA20502 3.50
Italian_Tuscan:NA20506 3.20
French:France35 3.05
Greek_Crete:B_Crete-1 2.45
Samaritan:GSM537033 1.45

Italian Tuscan: 42% (some of it Greek-derived, and Levantine-derived)
Cretan-like Greek: 21.5%
Levantine: 22%
French (aka Norman): ~14.5%

Now, let's try Italian Bergamo to cancel out any significant Levantine or Greek admixture which might exist in modern Tuscan Italians to some degree. Also, tried to add some Nordic populations to maybe emphasis Norman admixture:

[1] "distance%=0.6227 / distance=0.006227"

Italian_South

Greek_Crete:B_Crete-2 33.25
Italian_Bergamo:HGDP01151 15.95
Samaritan:168723 12.10
Italian_Bergamo:HGDP01153 9.20
Italian_Bergamo:HGDP01147 8.25
Greek_Crete:B_Crete-1 7.60
Greek_Crete:NA17376 4.40
French:France16 3.75
French:France9 3.55
Samaritan:GSM537033 1.95

Cretan-like Greek: 45.25% (should be translated to ~40%)
Levantine: 14% (should be translated to ~19%)
Northern Italian: 33.4% (how much is German/Celtic?)
French: 7.3%

Regarding the Greek - seems like I was correct - using Bergamo Italians, which are further north and are more likely less admixed with Greeks than Tuscans, elevated the amount of Greek admixture.

Which fit this model:

"Overall the estimated Balkan and North Western European paternal contributions in South Italy and Sicily are about 63% and 26% respectively."
https://en.wikipedia.org/wiki/Genetic_history_of_Italy#Y-DNA_introduced_by_historical_immigration

And, as I said - S. Italians are a mixture of Greeks, Northern Europeans and Levantines - which is the reason why they cluster so close to Western Jews. If Cypriots would have been more mixed with Northern Europeans, I believe Ashkenazi Jews would have clustered with them.


EDIT: Thought about modelling with Sardinians, as they represent older European genetic "form":

[1] "distance%=0.6307 / distance=0.006307"

Italian_South

Greek_Crete:B_Crete-2 43.90
French:France38 11.50
Sardinian:HGDP01073 10.15
French:France1 9.95
Greek_Crete:B_Crete-1 8.10
Samaritan:S_Samaritan-1 7.90
Samaritan:168723 5.70
Swedish:Sweden5 1.50
Greek_Crete:NA17376 1.30

Again, seem like Cretan-like Greek get's extremely high proportions.


Can you try to model Crete as other populations?

What is interesting is autosomally and on PCA plots, Cretans themselves are very close to southern Italians (there are Sicilians on GEDMatch who score Crete first). So I would expect Cretans to be modeled with Italian populations, also.

Interesting that Samaritans are a good proxy for the Levantine input in Sicily. This would make sense if Samaritans are close to Phoenicians.

Erikl86
06-10-2018, 04:23 PM
Can you try to model Crete as other populations?

What is interesting is autosomally and on PCA plots, Cretans themselves are very close to southern Italians (there are Sicilians on GEDMatch who score Crete first). So I would expect Cretans to be modeled with Italian populations, also.

Interesting that Samaritans are a good proxy for the Levantine input in Sicily. This would make sense if Samaritans are close to Phoenicians.

That is not surprising at all. S. Italy and most of Sicily have been part of the Greek world from the Iron Age (if not late Bronze Age) all the way to ~1100 AD.

Crete is one of the least Slavic-admixed Greek Island with significant Levantine ancestry. The two people - Cretans and S. Italians / Sicilians, should be similar.

Smaritans have been proven again and again to be the closest modern population to BA Canaanites, which were basically same as Phoenicians.

Sikeliot
06-10-2018, 04:35 PM
Crete is one of the least Slavic-admixed Greek Island with significant Levantine ancestry. The two people - Cretans and S. Italians / Sicilians, should be similar.


Try to model Crete with non-Greek populations? I want to see how they turn out.

Erikl86
06-10-2018, 04:44 PM
Try to model Crete with non-Greek populations? I want to see how they turn out.

I will, but there's no point in modelling them with S. Italians, since South Italians are already heavily Greek, as I've shown and as historically been so.
I would try to model Cretans with Mycenaean, Minoans, Slavic and Levant.

Sikeliot
06-10-2018, 04:48 PM
I will, but there's no point in modelling them with S. Italians, since South Italians are already heavily Greek, as I've shown and as historically been so.
I would try to model Cretans with Mycenaean, Minoans, Slavic and Levant.

So are you saying the "Cretan" part of your modeling for southern Italian is ancient Greek ancestry? But if Cretans also have Levantine, couldn't using Cretans to model them, mask some Levantine ancestry?

I would put French and Sardinians in there too just to see how they compare to the Sicilians.

I would take Minoans/Mycenaeans out. I think we should see how Cretans come out with modern populations, so yes, I do think you should put south Italians in there too, and different Italian groups.

Sikeliot
06-11-2018, 02:57 AM
I re-read Sarno et al and maybe the way that makes more sense to view the lack of IBD sharing between Greece and southern Italy is that Calabria and east-central Sicily (really all of Sicily except Trapani, in that study) have just not absorbed any mainland Greek admixture in the last 1500 years since mainland Greeks stopped being a genetically eastern Mediterranean population, but Apulia, Lucania, and Trapani have, and minor gene flow between mainland Greece and the Aegean islands has occurred. Therefore if there is Greek admixture in Sicily and Calabria it simply came from a very ancient population that there is nothing resembling in modern day mainland Greece.

Erikl86
06-11-2018, 06:47 AM
I re-read Sarno et al and maybe the way that makes more sense to view the lack of IBD sharing between Greece and southern Italy is that Calabria and east-central Sicily (really all of Sicily except Trapani, in that study) have just not absorbed any mainland Greek admixture in the last 1500 years since mainland Greeks stopped being a genetically eastern Mediterranean population, but Apulia, Lucania, and Trapani have, and minor gene flow between mainland Greece and the Aegean islands has occurred. Therefore if there is Greek admixture in Sicily and Calabria it simply came from a very ancient population that there is nothing resembling in modern day mainland Greece.

I think you are confusing terms here.

There is IBD between Greece and southern Italy. However, the highest IBD is between the most "Greek" areas of Greece - that is the more isolated Aegean and Adriatic Islands - than mainland Greece.

The entire area - S. Italy and Sicily, was Greek, by population, by language etc. for almost 1500 years. Most of the Greek have settled and came from Greece - yes even mainland Greece - long before the Salvic migration to the Balkans in the 6th-7th centuries AD.

I'll use visual aids to help communicate this.

PART I

1. Mycenaean expansion and trade routes, including S. Italy and Sicily, around the 13th century BC (more than 3200 years ago - during the Bronze Age !):

23862

1. Major Greek colonization of the Italians peninsula, starting 8th century BC, until the Hellenistic period (starting ~300 BC):

23863

23864

This later on, formed what is known as "Magna Graecia" by the Romans, and was crucial to the formation of both Roman and Etruscan civilizations, that got their writing systems from those Greeks areas in S. Italy.

Check out the different dialects - many of them came from mainland Greece at the time:

23865

In parallel to Greek settlement and colonization of the West Mediterranean, there was of course also massive, similar colonization and influence in the East. It began long before the Hellenistic time, as you can see in the maps above from the 8th century BC and mainly in Anatolia, but it accelerated immensely, starting with the Greek Macedonian empire of Alexander the Great:

23866

Erikl86
06-11-2018, 07:09 AM
PART II


3. After his death, the empire broke apart, and two major Greek Empires were created in the East and in Egypt. These Greek empires got a lot of additional Greek settlers, as described by historical records:


The history of Greeks in Syria traditionally begins with Alexander the Great's conquest of the Persian Empire. In the aftermath of Alexander's death, his empire was divided into several successor states, and thus ushering in the beginning of the Hellenistic Age. For the Levant and Mesopotamia, it meant coming under the control of Seleucus I Nicator and the Seleucid Empire. The Hellenistic period was characterized by a new wave of Greek colonization.[5] Ethnic Greek colonists came from all parts of the Greek world, not, as before, from a specific "mother city".[6] The main centers of this new cultural expansion of Hellenism in the Levant were cities like Antioch, and the other cities of the Tetrapolis Seleukis. The mixture of Greek-speakers gave birth to a common Attic-based dialect, known as Koine Greek, which became the lingua franca throughout the Hellenistic world.

The Seleucid Empire was a major empire of Hellenistic culture that maintained the pre-eminence of Greek customs in which a Greek political elite dominated, in newly founded urban areas.[7][8][9][10] The Greek population of the cities who formed the dominant elite were reinforced by emigration from Greece.[7][8] The creation of new Greek cities were aided by the fact that the Greek mainland was overpopulated and therefore made the vast Seleucid Empire ripe for colonization. Apart from these cities, there were also a large number of Seleucid garrisons (choria), military colonies (katoikiai) and Greek villages (komai) which the Seleucids planted throughout the empire to cement their rule.
Source: https://en.wikipedia.org/wiki/Greeks_in_Syria#Hellenistic_Age

Got it? There was massive Greek colonization to the East, as well.

This is a map of the Seleucid Empire, which existed from 300 BC to 63 BC:

23868

It spoke Greek, and was colonized by ethnic Greeks.

https://en.wikipedia.org/wiki/Seleucid_Empire

It fell to the Roman and Persian Empires around 63 BC:

23869

But the area continued to be Greek. Especially the areas on the coast:

Dark blue - regions with Greek majority; Pale blue - regions that spoke Greek and were highly Hellenized
23867

4. Even during the Roman Empire, those areas were still Greek, as Greek and Latin were both the official languages of the Roman Empire, and the Romans had huge respect and admiration to the Greek culture - after all they were heavily influenced by that culture even before they became an empire. This is the linguistic division of the Unified Roman Empire (red- latin, green - greek):

23870

As you can see - Greek was still spoken in S. Italy and Sicily. It was part of the Greek world.

5. Finally, during Byzantine period - this area remained under their reign all the way to the 11th century:

23871


So there you have it - from around 3200 BC all the way to ~1100 AD - almost 2300 years, Greeks and before them Mycenaeans have been colonizing, settling and living in S. Italy and Sicily (and of course, the Aegean sea). This is was all one world.

There was no major Greek migration to Italy after the 8th century AD, only minor settlements later on, sometimes re-enforced the Byzantine Empires. And of course, after the fall of Constantinople in 1453, there were Greeks coming - from mainland Greece as well - to Italy, ushering the Renaissance period. But overall, most of the Greeks in Italy came long, long before that.

Erikl86
06-11-2018, 07:10 AM
PART III

Slavs arrived to mainland Greece around 600-700 AD. There is no major Slavic influence in most of the Greek islands, and of course Cyprus, which were much closer to Greece mainland. So obviously, the chances that there would be a major Slavic influence in Italy and Sicily, are even smaller:

23872

Erikl86
06-11-2018, 07:15 AM
So to summarize it, the reason why there is no point in modelling Greeks with southern Italians, is the same reason why you wouldn't model your parents using your samples. Greeks colonized S. Italy and Sicily, not the other way around.

Seabass
06-11-2018, 09:00 AM
That is even more preposterous as there is no chance Italkim are just 30% Levantines and 70 Tuscans. Even if the distance is better.

And now, instead of Levant BA, they should be pretty close to Cypriots, as basically they should lack the French or Moroccan admixture seen in Sephardic Jews:


Bulgarian/Turkish Jews are the closest Jewish group to Italian Jews and seem to show only the tiniest bit more potential for Berber admixture. I havn't found any compelling evidence that Bulgarian/Turkish Jews have any more say Iberian admixture then Italian Jews even though logically they should. Italian Jews autosomally closer to both Near Eastern and Southern European populations then Bulgarian/Turkish Jews despite not being as outbred as the latter because I suspect Bulgarian/Turkish Jews are more genetically diverse, ie Berber and African admixture tends to have this effect on West-Eurasian populations. Saying that though, you are aware Bulgarian/Turkish Jews are a tiny bit more 'Near Eastern' then Italian Jews? Ie shift closer to the Near East on both Global and Eurasia PCA's. Given that, how would we explain the potential for minor Iberian admixture being present in Bulgarian/Turkish Jews yet absent in Italian Jews? The results wouldn't make sense unless these below scenarios are a reality:

* Italian Jews went on to acquire a significant amount of gentile Italian admixture after say the Sephardic/Ashkenazi Jews branched off from them.
* Bulgarian/Turkish Jews following 1492 AD begun absorbing many Romaniote Jews all over Greece and Turkey who were possibly more 'Near Eastern' genetically by this time then Italqim?
* Frequent migrations between the Maghreb and Spain meant the tiny amount more Berber admixture Bulgarian/Turkish Jews possibly recieved compared to Italqim reduced both their autosomal European hunter gatherer and Steppe kind of ancient admixtures. Even if so, doesn't explain why Bulgarian/Turkish Jews show more Caucasus Hunter Gatherer then Italqim.


Philistines should be available soon, I'm unsure why we don't have non-Jewish Roman samples yet.

Are there any rumors when the Ashkelon genomes should be available to the public?

There are a few likely non-Jewish Roman samples. There are three from Egypt. (Egyptian_mummy in G25) A Roman soldier from Yorkshire England who is likely from inland Egypt or Jordan maybe. (England_Roman_outlier in G25) There is also a Roman from Germany who genetically appears very similar to the Southern French and Basque. (Germany_Roman in G25)

The model below emphasises my point about Euro/Mediterranean Jews differentiating from one another a little. The Iberomaursian is probably deflating the Natufian in Moroccan Jews a little. Another important observation is even though Cypriots appear more 'generic Levantine' then Western Jews, going off those Natufian scores below Jews could well have more Southern Levantine DNA then Cypriots for all I know. Euro/Mediterranean Jews seem to score over half the 'Natufian' and Neolithic Iranian components Samaritans score. Bronze Age Aegeans and Anatolians score very little 'Natufian' as do Abruzzo-Italians (the theoretical source for gentile admixture)


[1] "distance%=1.8004 / distance=0.018004"

Mycenaean

Barcin_N 70.65
Sintashta_MLBA 11.90
CHG 9.85
Ganj_Dareh_N 5.50
Natufian 1.50
Yoruba 0.50
West_Siberia_N 0.10
EHG 0.00
Dinka 0.00
WHG 0.00
Iberomaurusian 0.00

[1] "distance%=2.3591 / distance=0.023591"

Balkans_IA

Barcin_N 65.2
Sintashta_MLBA 26.6
CHG 4.4
Ganj_Dareh_N 3.7
Natufian 0.0
EHG 0.0
Yoruba 0.0
Dinka 0.0
WHG 0.0
West_Siberia_N 0.0
Iberomaurusian 0.0


[1] "distance%=2.273 / distance=0.02273"

Anatolia_BA

Barcin_N 66.60
CHG 15.95
Ganj_Dareh_N 14.40
Natufian 3.05
EHG 0.00
Yoruba 0.00
Dinka 0.00
Sintashta_MLBA 0.00
WHG 0.00
West_Siberia_N 0.00
Iberomaurusian 0.00


[1] "distance%=3.4105 / distance=0.034105"

Anatolia_EBA:MA2212

Barcin_N 59.1
Ganj_Dareh_N 17.2
CHG 13.1
Natufian 10.6
Sintashta_MLBA 0.0
EHG 0.0
Dinka 0.0
Yoruba 0.0
WHG 0.0
West_Siberia_N 0.0
Iberomaurusian 0.0

[1] "distance%=3.4548 / distance=0.034548"

Samaritan

Barcin_N 37.65
Natufian 30.25
Ganj_Dareh_N 22.95
CHG 9.15
EHG 0.00
Yoruba 0.00
Dinka 0.00
Sintashta_MLBA 0.00
WHG 0.00
West_Siberia_N 0.00
Iberomaurusian 0.00



[1] "distance%=2.4064 / distance=0.024064"

Cypriot

Barcin_N 46.85
Ganj_Dareh_N 16.30
Natufian 15.80
CHG 13.60
Sintashta_MLBA 7.45
EHG 0.00
Yoruba 0.00
Dinka 0.00
WHG 0.00
West_Siberia_N 0.00
Iberomaurusian 0.00


[1] "distance%=2.0613 / distance=0.020613"

Tunisian_Jew

Barcin_N 42.55
Natufian 20.80
Ganj_Dareh_N 13.85
Sintashta_MLBA 9.55
CHG 8.90
Iberomaurusian 3.75
Yoruba 0.60
EHG 0.00
Dinka 0.00
WHG 0.00
West_Siberia_N 0.00

[1] "distance%=2.2291 / distance=0.022291"

Moroccan_Jew

Barcin_N 42.75
Sintashta_MLBA 18.20
Natufian 14.35
Ganj_Dareh_N 12.75
Iberomaurusian 6.20
CHG 5.75
EHG 0.00
Yoruba 0.00
Dinka 0.00
WHG 0.00
West_Siberia_N 0.00

[1] "distance%=2.1322 / distance=0.021322"

Sephardic_Jew

Barcin_N 43.65
Sintashta_MLBA 17.75
Natufian 16.05
Ganj_Dareh_N 13.50
CHG 8.10
Iberomaurusian 0.95
EHG 0.00
Yoruba 0.00
Dinka 0.00
WHG 0.00
West_Siberia_N 0.00

[1] "distance%=1.5015 / distance=0.015015"

Italian_Jew

Barcin_N 44.1
Sintashta_MLBA 19.6
Natufian 17.4
Ganj_Dareh_N 11.6
CHG 7.2
EHG 0.0
Yoruba 0.0
Dinka 0.0
WHG 0.0
West_Siberia_N 0.0
Iberomaurusian 0.0

[1] "distance%=1.9077 / distance=0.019077"

Ashkenazi_Jew

Barcin_N 38.2
Sintashta_MLBA 27.2
Natufian 16.6
Ganj_Dareh_N 11.1
CHG 6.9
EHG 0.0
Yoruba 0.0
Dinka 0.0
WHG 0.0
West_Siberia_N 0.0
Iberomaurusian 0.0

[1] "distance%=1.4229 / distance=0.014229"

Maltese

Barcin_N 43.90
Sintashta_MLBA 27.80
Natufian 12.05
Ganj_Dareh_N 7.85
CHG 6.30
Iberomaurusian 1.45
Yoruba 0.65
EHG 0.00
Dinka 0.00
WHG 0.00
West_Siberia_N 0.00

[1] "distance%=1.4703 / distance=0.014703"

Italian_Abruzzo

Barcin_N 47.20
Sintashta_MLBA 32.95
CHG 7.25
Natufian 6.75
Ganj_Dareh_N 5.85
EHG 0.00
Yoruba 0.00
Dinka 0.00
WHG 0.00
West_Siberia_N 0.00
Iberomaurusian 0.00

Claudio
06-11-2018, 09:16 AM
I've actually addressed this study in the opening post of this thread:



Basically, my opinion is that it's not that Ashkenazim mixed with Italians only after 500 AD, it's that Italians have begun absorbing Greeks from S. Italy and Sicily en-mass starting that time, and being that Ashkenazim, according to my suggestion, are about 50% admixed with Aegeans, this is when a very similar admixture from non-Jewish Greco-Italians would start appearing among the rest of the Italian population. The time period fits when Byzantines lost for good big parts of Greek S. Italy, a process which followed with their Latinization and absorption in Latin-speaking Italy, and ended mostly around the turn of the millennium, with the Byzantines losing all of their territories in the Italian peninsula. The time frame fits, it seems.

Not gonna pretend I have the same knowledge as you guys concerning running Gedmatch programs and modeling populations etc.
But the more I look into your theory from a historical and geographical view point it does seem possible.
The Study that estimated the timing of Southern euro admix within the Jewish population?
Have you thought of investigating how it was conducted?
What the study actually considers Southern euro Admixture etc?
The devil may be in the details as they say?

Sikeliot
06-11-2018, 11:04 AM
So to summarize it, the reason why there is no point in modelling Greeks with southern Italians, is the same reason why you wouldn't model your parents using your samples. Greeks colonized S. Italy and Sicily, not the other way around.

Actually it did happen the other way around because after some of the Slavs from the Peloponnese were expelled (some were also assimilated), the Byzantines repopulated it with southern Italians. However, there was no southern Italian influence in northern or central Greece and a much higher number of Slavs, so that explains why those regions are not genetically similar to southern Italy.


I think you are confusing terms here.

There is IBD between Greece and southern Italy. However, the highest IBD is between the most "Greek" areas of Greece - that is the more isolated Aegean and Adriatic Islands - than mainland Greece.

The entire area - S. Italy and Sicily, was Greek, by population, by language etc. for almost 1500 years. Most of the Greek have settled and came from Greece - yes even mainland Greece - long before the Salvic migration to the Balkans in the 6th-7th centuries AD.


What I am saying is that in the period of time since Greece has had Slavic influence, there is no evidence in DNA of Greeks from the mainland settling or contributing to the DNA of Calabria or most of Sicily, because there is no DNA signal in those regions suggesting any infusion of NE European DNA. Therefore, any Greek ancestry either a) came from a part of Greece with no Slavic that was already similar to southern Italy in recent times, or b) is very ancient and predates Slavic DNA and comes from a group that are not genetically similar to any modern mainland Greeks.

My question for you would be, how much of southern Italian DNA would you say comes from pre-Greek Italic groups? We already established there is somewhere like 20% Levantine that arrived independently from everything else, so what about Italic?

And what is your best estimate for the actual amount of Slavic present in Greece? It must be significant in some regions because it does influence phenotype. Any group of modern mainland Greeks, for the most part, could not be mistaken as a group for southern Italian nor Dodecanese islanders.


Not gonna pretend I have the same knowledge as you guys concerning running Gedmatch programs and modeling populations etc.
But the more I look into your theory from a historical and geographical view point it does seem possible.
The Study that estimated the timing of Southern euro admix within the Jewish population?
Have you thought of investigating how it was conducted?
What the study actually considers Southern euro Admixture etc?
The devil may be in the details as they say?

It could make sense to me if we are to believe that Jews mixed with Greek-admxied people IN ITALY but there is nothing to suggest the admixture happened within the borders of modern Greece. As they moved up Italy they would have encountered people who were Greek-speaking. But I absolutely do not believe the admixture occurred while the Jews were still in the Levant.

Sikeliot
06-11-2018, 11:08 AM
Are there any rumors when the Ashkelon genomes should be available to the public?

No.
I have wondered and no one answered -- do you expect them to be close to any modern populations? Or would they be, like Mycenaeans and Bronze Age Sicilians, not quite akin to anyone and have large distances on calculators with their top 3 matches?

giorgio
06-11-2018, 11:52 AM
Actually it did happen the other way around because after some of the Slavs from the Peloponnese were expelled (some were also assimilated), the Byzantines repopulated it with southern Italians. However, there was no southern Italian influence in northern or central Greece and a much higher number of Slavs, so that explains why those regions are not genetically similar to southern Italy.




What I am saying is that in the period of time since Greece has had Slavic influence, there is no evidence in DNA of Greeks from the mainland settling or contributing to the DNA of Calabria or most of Sicily, because there is no DNA signal in those regions suggesting any infusion of NE European DNA. Therefore, any Greek ancestry either a) came from a part of Greece with no Slavic that was already similar to southern Italy in recent times, or b) is very ancient and predates Slavic DNA and comes from a group that are not genetically similar to any modern mainland Greeks.

My question for you would be, how much of southern Italian DNA would you say comes from pre-Greek Italic groups? We already established there is somewhere like 20% Levantine that arrived independently from everything else, so what about Italic?

And what is your best estimate for the actual amount of Slavic present in Greece? It must be significant in some regions because it does influence phenotype. Any group of modern mainland Greeks, for the most part, could not be mistaken as a group for southern Italian nor Dodecanese islanders.



It could make sense to me if we are to believe that Jews mixed with Greek-admxied people IN ITALY but there is nothing to suggest the admixture happened within the borders of modern Greece. As they moved up Italy they would have encountered people who were Greek-speaking. But I absolutely do not believe the admixture occurred while the Jews were still in the Levant.

Some South Italians even for today don't consider themselves as Italian nor their ancestors spoke any Italian until the forced assimilation started in the 19th century and so. A large portion of Calabria and Sicily has always been Greek speaking and their Greek is closer to Ancient Hellenic dialect compared to modern Greek giving the suggestion that modern Greek language in Greece hasn't been spoken continuously but a large portion of it's population was hellenized from the Middle ages to recent times (Think of Thessaly which was called as Vlachia and their inhabitants barely spoke Greek nor identified as such) I believe Hellenic ancestry is possibly stronger in South Italy than in a lot parts of the Mainland territories where it's natives are largely hellenized (Vlachs, Pontians, Fyromanians, Albanians/Arvanites, Bulgars even some Turks and who knows what else) If anything South Italians carry a far higher DNA overlapping with early Hellenes (Myceaneans and Minoans) even if they weren't Greek speaker from the beginning (at least not entirely) they were akin to early Hellenic people and most likely came from the same pre Hellenic "Sea people like" branch who lately mixed with some Italic and some "Levantine" stock of people who came via the Sea but I believe even Mainland Greece has some recent Anatolian, Levant Geneflow received but their high Slavic and Balkanic ancestry balances it out, where as South Italians barely received any Steppe or Continental European settlements since the Middle ages (given how many German speakers moved across the whole European peninsula it's not rare today to find people of German ancestry even in Andalusia)

Also I think a portion of Calabrese and some Sicilians have assimilated Jewish ancestry that can make them more Levant shifted but this doesn't account for every Sicilian, South Italian. There are Sicilian kits who cluser with Southern Mainlanders and far removed from most Jewish samples. Actually South Italy was once a melting plot and those from Aberesche founded towns will certainly be different from those of Greek or ex Greek speaking towns. While I wouldn't doubt in regions where the local Marrano, Italkim community was stronger has some obvious Jewish presence (given some Calabrians are even more Levant shifted than most European Jews like these Gedmatch KITs A651078, A017239 from Calabria (not sure which town it is)

Erikl86
06-11-2018, 11:55 AM
Given that, how would we explain the potential for minor Iberian admixture being present in Bulgarian/Turkish Jews yet absent in Italian Jews?


Well we can explain it for the fact that Bulgarian and Turkish Jews are mainly Sephardic in origins, while Italkim have never been in Spain.



* Italian Jews went on to acquire a significant amount of gentile Italian admixture after say the Sephardic/Ashkenazi Jews branched off from them.
* Bulgarian/Turkish Jews following 1492 AD begun absorbing many Romaniote Jews all over Greece and Turkey who were possibly more 'Near Eastern' genetically by this time then Italqim?
* Frequent migrations between the Maghreb and Spain meant the tiny amount more Berber admixture Bulgarian/Turkish Jews possibly recieved compared to Italqim reduced both their autosomal European hunter gatherer and Steppe kind of ancient admixtures. Even if so, doesn't explain why Bulgarian/Turkish Jews show more Caucasus Hunter Gatherer then Italqim.


I agree with all of those assumptions and assume so as well.





Are there any rumors when the Ashkelon genomes should be available to the public?

No, I have no idea when will it be published.


There are a few likely non-Jewish Roman samples. There are three from Egypt. (Egyptian_mummy in G25) A Roman soldier from Yorkshire England who is likely from inland Egypt or Jordan maybe. (England_Roman_outlier in G25) There is also a Roman from Germany who genetically appears very similar to the Southern French and Basque. (Germany_Roman in G25)

Yes, but none of them are Romans from Rome. We need Roman-time Italian samples (hopefully not S. Italians as those would have been very close to Greeks already by Roman times).


The model below emphasises my point about Euro/Mediterranean Jews differentiating from one another a little. The Iberomaursian is probably deflating the Natufian in Moroccan Jews a little. Another important observation is even though Cypriots appear more 'generic Levantine' then Western Jews, going off those Natufian scores below Jews could well have more Southern Levantine DNA then Cypriots for all I know. Euro/Mediterranean Jews seem to score over half the 'Natufian' and Neolithic Iranian components Samaritans score. Bronze Age Aegeans and Anatolians score very little 'Natufian' as do Abruzzo-Italians (the theoretical source for gentile admixture)

Well Natufian is Neolithic Levantine, which is different than Bronze Age Levantine, mainly as a result of CHL Iranian migrations that happened 6000 years ago. But yes, it would make sense that Bronze Age Anatolian and Aegean would have very little Neolithic Levantine in them.

Seabass
06-11-2018, 12:35 PM
Well we can explain it for the fact that Bulgarian and Turkish Jews are mainly Sephardic in origins, while Italkim have never been in Spain.


How Romaniote ancestrally many Jews from Turkey and Bulgaria today really are is difficult to gauge. For a good couple of centuries in Istanbul, a census of Jews there suggested it was 50/50 Sephardic and Romaniote. There was initially resistance to intermarry, but it seems like they were eventually assimilated into the Sephardic population. It was the same story for Greece, however I think more Yevanic-speaking Jews survived in Greece up until sadly the holocaust. Nazi's literally genocided the Yevanic-speaking Jews. You would be lucky to maybe find a minority today in Greece and maybe New York. Based on the Azmon study, Greek Jews (pretty sure they were Sephardic) were closest to Turkish Jews which isn't exactly surprising.

From memory I THINK i did have a gedmatch ID of someone who claimed their mother was full Romaniote partially from Greece and Turkey. I'll have to try fish from it, but I remembered her results were the same as Sephardic Jews, the same way Italqim's basically are.

Those sample of Italian Jews, how do we know they are Italqim? If they are from the Atzmon study they just come from the 'Italian Jewish community of Rome'. They could very well have absorbed a lot of Sephardic emigrees.

Atzmon study below.

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3032072/pdf/main.pdf

Erikl86
06-11-2018, 01:03 PM
Actually it did happen the other way around because after some of the Slavs from the Peloponnese were expelled (some were also assimilated), the Byzantines repopulated it with southern Italians. However, there was no southern Italian influence in northern or central Greece and a much higher number of Slavs, so that explains why those regions are not genetically similar to southern Italy.

There was also Venetian rule on Aegean Islands for several centuries - yet the populations didn't mix because they were already separated religiously at the time, plus the local Greeks despised their Catholic overlords and rebelled against them. Italian admixture among Greeks is very small and have little to no significance, much smaller than the other way around.



What I am saying is that in the period of time since Greece has had Slavic influence, there is no evidence in DNA of Greeks from the mainland settling or contributing to the DNA of Calabria or most of Sicily, because there is no DNA signal in those regions suggesting any infusion of NE European DNA. Therefore, any Greek ancestry either a) came from a part of Greece with no Slavic that was already similar to southern Italy in recent times, or b) is very ancient and predates Slavic DNA and comes from a group that are not genetically similar to any modern mainland Greeks.

The highlighted part is probably the case.


My question for you would be, how much of southern Italian DNA would you say comes from pre-Greek Italic groups? We already established there is somewhere like 20% Levantine that arrived independently from everything else, so what about Italic?
Davidsky answered that in his Eurogenes project blog (and I've posted it several times now):


Greek (from Thessaloniki):
Iran_ChL 0.090±0.071
Mycenaean 0.478±0.103
Slav_Bohemia 0.432±0.077
P-value 0.461783732
chisq 12.820
Full output

Italian_Bergamo
Anatolia_BA 0.239±0.057
Iceman_MN 0.332±0.054
Unetice 0.429±0.030
P-value 0.764439946
chisq 9.112
Full output

Sicilian_East
Bell_Beaker_Germany 0.222±0.077
England_Roman_outlier 0.210±0.134
Mycenaean 0.567±0.163
P-value 0.504442682
chisq 12.285
Full output

Sicilian_West
England_Roman_outlier 0.216±0.121
Mycenaean 0.503±0.135
Unetice 0.281±0.056
P-value 0.808464904
chisq 8.516
Full output



And it roughly fits my own model from yesterday:


[1] "distance%=0.6227 / distance=0.006227"

Italian_South

Greek_Crete:B_Crete-2 33.25
Italian_Bergamo:HGDP01151 15.95
Samaritan:168723 12.10
Italian_Bergamo:HGDP01153 9.20
Italian_Bergamo:HGDP01147 8.25
Greek_Crete:B_Crete-1 7.60
Greek_Crete:NA17376 4.40
French:France16 3.75
French:France9 3.55
Samaritan:GSM537033 1.95

Cretan-like Greek: 45.25% (should be translated to ~40%)
Levantine: 14% (should be translated to ~19%)
Northern Italian: 33.4% (how much is German/Celtic?)
French: 7.3%

So I would say, in Siciliy, probably 20-30% derives from BA Italian-like people, and 50-60% is Aegean Greek, and +20% is Levantine and North African.



And what is your best estimate for the actual amount of Slavic present in Greece? It must be significant in some regions because it does influence phenotype. Any group of modern mainland Greeks, for the most part, could not be mistaken as a group for southern Italian nor Dodecanese islanders.


I wouldn't know, and I wouldn't trust phenotype at all, just genetics. My sister looks Mediterranean, with tanned skin tone, curly brown-black hair and brown eyes, while I'm a red hair, with smooth hair, blue eyed, small-nosed pale skin guy.

Yet we have the same parents.



It could make sense to me if we are to believe that Jews mixed with Greek-admxied people IN ITALY but there is nothing to suggest the admixture happened within the borders of modern Greece. As they moved up Italy they would have encountered people who were Greek-speaking. But I absolutely do not believe the admixture occurred while the Jews were still in the Levant.

I don't think so. I think that the admixture with Greeks happened even before the Levantine Judeans set foot outside of the Levant, back in Hellenistic times and perhaps even as a result of Philistine admixture. Remember that Seleucid Empire map I've shown you and Greek settlements areas in the Levant? I think the Jews which have settled in Italy, most likely Hellenistic Jews, were already pretty mixed with Greeks (and they also absorbed Greek converts on their way West so they would be even more admixed with them).

I ran again Sephardic Jews against Cypriots and Southern Italians on Global25.

Check it out:


Control group - Sephardic_Jew average vs. Sephardic Jews:

[1] "distance%=0.338 / distance=0.00338"

Sephardic_Jew

Sephardic_Jew:sephardic14bul 56.6
Sephardic_Jew:sephardic16bul 43.4

Then,

Cypriots + Southern French (without any North African population):

[1] "distance%=0.8434 / distance=0.008434"

Sephardic_Jew

Cypriot:CYP2 34.6
Cypriot:Cyprus24AJ19 32.4
French_South:SouthFrench1112 19.9
Cypriot:CYP5 8.8
Cypriot:Cyprus21AJ19 2.4
Cypriot:Cyprus22AJ19 1.9



Cypriots + Southern French + Mozabite Berbers (North Africans with almost no Arabic admixture):

[1] "distance%=0.5482 / distance=0.005482"

Sephardic_Jew

Cypriot:CYP2 28.10
Cypriot:Cyprus24AJ19 18.90
Cypriot:CYP5 16.40
French_South:SouthFrench3326 9.80
Mozabite:HGDP01266 9.25
Cypriot:Cyprus21AJ19 8.00
French_South:SouthFrench1112 7.15
Cypriot:CYP19 2.40

Cypriot averages + Southern French averages + Mozabite berbers averages:

[1] "distance%=0.9262 / distance=0.009262"

Sephardic_Jew

Cypriot 76.2
French_South 12.7
Mozabite 11.2

Now, Cypriots themselves, at single distance, are further away from Sephardic Jews than S. Italians and Sicilians:

S. Italians:

[1] "1. CLOSEST SINGLE ITEM DISTANCES"
Italian_South:ITS7 Italian_South:BEL57
0.01830759 0.01932739
Italian_South:ITS4 Italian_South:ITS2
0.01987400 0.02018242

Sicilians:

[1] "1. CLOSEST SINGLE ITEM DISTANCES"
Sicilian_West:WestSicilian7H Sicilian_West:WestSicilian4H
0.01469625 0.01772354
Sicilian_East:EastSicilian5H Sicilian_East:EastSicilian8H
0.01811033 0.01859198
Sicilian_East:EastSicilian2H Sicilian_West:WestSicilian10H
0.01907878 0.02196251


Cypriots:

[1] "1. CLOSEST SINGLE ITEM DISTANCES"
Cypriot:Cyprus24AJ19 Cypriot:CYP2 Cypriot:CYP19
0.01724215 0.01976406 0.02018351
Cypriot:Cyprus22AJ19 Cypriot:Cyprus21AJ19 Cypriot:Cyprus13AJ19
0.02280785 0.02331300 0.02529158
Cypriot:CYP5 Cypriot:Cyprus2AJ19
0.02713252 0.02863833



So, let's replace Cypriots with S. Italians and Sicilians. Watch what happens, both to the distance, and also to the South French component:


Southern Italians + Southern French + Mozabite berbers:

[1] "distance%=1.0907 / distance=0.010907"

Sephardic_Jew

Italian_South:ITS2 35.00
Italian_South:BEL57 32.55
Italian_South:ITS7 18.00
Mozabite:HGDP01268 8.90
Mozabite:HGDP01273 5.55
French_South:SouthFrench1112 0.00
French_South:SouthFrench1323 0.00
French_South:SouthFrench3068 0.00
French_South:SouthFrench3326 0.00
French_South:SouthFrench3947 0.00
French_South:SouthFrench3951 0.00
French_South:SouthFrench4018 0.00
Mozabite:HGDP01253 0.00
Mozabite:HGDP01254 0.00
Mozabite:HGDP01255 0.00
Mozabite:HGDP01256 0.00
Mozabite:HGDP01257 0.00
Mozabite:HGDP01258 0.00
Mozabite:HGDP01259 0.00
Mozabite:HGDP01260 0.00
Mozabite:HGDP01262 0.00
Mozabite:HGDP01263 0.00
Mozabite:HGDP01264 0.00
Mozabite:HGDP01265 0.00
Mozabite:HGDP01266 0.00
Mozabite:HGDP01272 0.00
Mozabite:HGDP01274 0.00
Mozabite:HGDP01275 0.00
Mozabite:HGDP01276 0.00
Mozabite:HGDP01277 0.00
Mozabite:HGDP01278 0.00
Mozabite:HGDP01279 0.00
Mozabite:HGDP01280 0.00
Italian_South:ITS4 0.00


Sicilians + Southern French + Mozabite berbers:

[1] "distance%=1.0564 / distance=0.010564"

Sephardic_Jew

Sicilian_East:EastSicilian8H 29.45
Sicilian_East:EastSicilian5H 27.50
Sicilian_East:EastSicilian2H 17.50
Sicilian_West:WestSicilian7H 14.30
Mozabite:HGDP01266 8.60
Mozabite:HGDP01268 2.20
Mozabite:HGDP01273 0.45
French_South:SouthFrench1112 0.00
French_South:SouthFrench1323 0.00
French_South:SouthFrench3068 0.00
French_South:SouthFrench3326 0.00
French_South:SouthFrench3947 0.00
French_South:SouthFrench3951 0.00
French_South:SouthFrench4018 0.00
Mozabite:HGDP01253 0.00
Mozabite:HGDP01254 0.00
Mozabite:HGDP01255 0.00
Mozabite:HGDP01256 0.00
Mozabite:HGDP01257 0.00
Mozabite:HGDP01258 0.00
Mozabite:HGDP01259 0.00
Mozabite:HGDP01260 0.00
Mozabite:HGDP01262 0.00
Mozabite:HGDP01263 0.00
Mozabite:HGDP01264 0.00
Mozabite:HGDP01265 0.00
Mozabite:HGDP01272 0.00
Mozabite:HGDP01274 0.00
Mozabite:HGDP01275 0.00
Mozabite:HGDP01276 0.00
Mozabite:HGDP01277 0.00
Mozabite:HGDP01278 0.00
Mozabite:HGDP01279 0.00
Mozabite:HGDP01280 0.00
Sicilian_West:WestSicilian10H 0.00
Sicilian_West:WestSicilian4H 0.00

Did you see what happened? First of all, by using Italians rather than Cypriots, the South French component disappears. The distance grows, and, in the case of Sicilians - it's mostly Eastern Sicilians who show up in the admixture, and West Sicilians show up at around ~14%.

Regarding the Sicilians - look at these maps:

23875
Phoenician time - around 8th century BC

23876
Late Hellenistic period - around 1st century BC

23877
Roman period

Did you see it? Most Greek colonization happened in the East. While of course Sicilians moved around, but the vast majority of Greeks in Sicily lived in the East. This is the most Greek region in Sicily.


Now, regarding the disappearance of Southern French, look how they can be constructed:

[1] "distance%=1.0463 / distance=0.010463"

French_South

French:France13 47.55
Italian_Bergamo:HGDP01151 32.35
French:France12 9.05
French:France15 7.90
Italian_Tuscan:NA20505 3.15

Southern French, when modeled between French and Italians, turn out to be ~35% Italian (mostly northern Italians, which would make sense). That would explain why South French disappear once Sephardic Jews are modeled with S. Italians - they are pretty close to French, especially Southern French, that the small amount just disappears.



The more I look at it, the more I see that the only reason why Western Jews cluster so close to Southern Italian and Sicilians, is that just like Western Jews, S. Italians and Sicilians are a mixture of Levantines, Aegean people and local population similar to French. Cypriots, which would model Western Jews much better being Aegean + Levantine, lack the additional French-like admixture so in single distance, they are further away.

Cascio
06-11-2018, 02:14 PM
Thats a ver high %.....considering the empire ruked about 13M

Maybe its a ficticous religous number since there was only jews and "pagans" at that tlme...pre-christians

The Roman Empire had a lot more than 13 million people.

Estimates run from 50 to 100 millions.

Erikl86
06-11-2018, 02:26 PM
One last thing.... I've tried admixing Cypriots + South French + Mozabites, plus different Italian groups. Look what happens:

With Sicilians:

[1] "distance%=0.4631 / distance=0.004631"

Sephardic_Jew

Sicilian_West:WestSicilian7H 20.35
Sicilian_East:EastSicilian5H 18.55
Cypriot:CYP5 13.50
Cypriot:CYP2 13.00
Cypriot:Cyprus24AJ19 9.70
Mozabite:HGDP01266 9.25
Sicilian_East:EastSicilian2H 5.15
Cypriot:Cyprus2AJ19 4.20
Cypriot:CYP19 3.70
Cypriot:Cyprus22AJ19 2.60

Sicilian total: ~44% (mostly East).
Cypriot total: ~47%
Mozabite: ~9%
South French: Disappeared (probably higher French-Norman admixture for Sicilians, that were ruled by them for 800 years).

With South Italians (which should be less Greek and less Levantine than Sicilians) :

[1] "distance%=0.5263 / distance=0.005263"

Sephardic_Jew

Cypriot:CYP2 20.95
Cypriot:Cyprus24AJ19 18.30
Italian_South:ITS7 14.90
Mozabite:HGDP01266 9.45
Italian_South:BEL57 8.40
French_South:SouthFrench1112 8.05
Cypriot:CYP5 8.05
Cypriot:CYP19 5.20
Cypriot:Cyprus2AJ19 3.50
Cypriot:Cyprus21AJ19 3.20

Total S. Italians: ~23%
Total Cypriot: ~59%
Mozabite: remained ~9%.

Southern French returned (8%) - and it's about half of what I'd get for it if I hadn't used Italians, remember:


Cypriots + Southern French + Mozabite Berbers (North Africans with almost no Arabic admixture):

[1] "distance%=0.5482 / distance=0.005482"

Sephardic_Jew

Cypriot:CYP2 28.10
Cypriot:Cyprus24AJ19 18.90
Cypriot:CYP5 16.40
French_South:SouthFrench3326 9.80
Mozabite:HGDP01266 9.25
Cypriot:Cyprus21AJ19 8.00
French_South:SouthFrench1112 7.15
Cypriot:CYP19 2.40

So, 8% vs. 17%. Probably less Norman admixture for S. Italians. I'd expect as we got north in Italy, it'll disappear again because Northern Italians are more similar to Southern French, as established already.

Now, let's go even more north, with Italians from Abruzzo, central Italy:

[1] "distance%=0.4344 / distance=0.004344"

Sephardic_Jew

Italian_Abruzzo:ItalyAbruzzo14 26.85
Cypriot:CYP2 11.90
Cypriot:Cyprus24AJ19 11.35
Italian_Abruzzo:ItalyAbruzzo15 9.55
Cypriot:CYP19 7.75
Cypriot:Cyprus2AJ19 6.80
Mozabite:HGDP01266 5.60
Mozabite:HGDP01275 5.55
Cypriot:Cyprus13AJ19 3.45
Italian_Abruzzo:ItalyAbruzzo16 3.15
Cypriot:CYP5 2.95
Cypriot:Cyprus21AJ19 2.80
Italian_Abruzzo:ItalyAbruzzo23 2.30


Total Abruzzo Italians: ~42%
Total Cypriots: ~47%
Mozabite: ~11%

But - S. French disappeared, just like in the case of Sicilians. The results are very similar to Sicilians indeed. Mozabite grew a little bit - but always remains around ~10% give or take. Again, some of the 42% of the Abruzzo Italian - corresponds to the S. French, remember.

Abruzzo should also still be pretty Greek admixed, because it borders S. Italy and used to be part of the Kingdom of Two Sicilies - a kingdom which brought together this region and all of S. Italy.

Let's go even further - to the Tuscans:

[1] "distance%=0.5331 / distance=0.005331"

Sephardic_Jew

Cypriot:CYP2 28.40
Cypriot:Cyprus24AJ19 21.95
Cypriot:CYP5 11.40
Mozabite:HGDP01266 9.55
French_South:SouthFrench1112 8.15
Italian_Tuscan:NA20508 8.10
Cypriot:Cyprus21AJ19 5.60
Italian_Tuscan:NA20506 4.70
French_South:SouthFrench3326 2.15


Total Tuscans: ~13%
Total Cypriot: ~67%
Total Mozabite: ~9.5%
Total S. French: ~10%


Well well.... what do you say to that? S. French returned - but it can almost be added to the Tuscan to get a very similar amount of S. French without Italians. I guess some percentage of Tuscan is also Greek, from some migration from the South to the North of Italy over the centuries, especially after the Latinization of the south, starting from ~1100 AD.

The tricky part is the disappearance and re-appearance of S. French component - but I think it can be explained as following:
Sicilians are mixed with Normans - French people. This admixture overlaps with S. French.
South Italians have much less Norman admixture, and are also different enough from North Italians for S. French to not overlap - so it appears again - but less than half.
Central Italians - are now closer to Tuscans, may still have some Norman admixture (not a lot), and are genetically closer to S. French - so it disappears again.
Tuscans - are much closer to S. French, but still different, and there is no Norman admixture, so it re-appears, but again - only a little bit more than half.

I predict, Bergamo would completely overlap with S. French, and would actually appear roughly the same percentage as S. French without Italian admixture - around 17%, plus a little bit higher on the account of Cypriot percentage, because perhaps there is still some very low admixture with Greeks from S. Italians migrating north - and also Renaissance Greeks settling in Northern Italy in the 15th-16th centuries (Venice and surrounding areas).


Now, last but not least, Bergamo Italians:

[1] "distance%=0.4776 / distance=0.004776"

Sephardic_Jew

Cypriot:CYP2 16.70
Italian_Bergamo:HGDP01155 14.15
Cypriot:CYP5 13.45
Cypriot:Cyprus21AJ19 12.40
Cypriot:Cyprus24AJ19 11.85
Italian_Bergamo:HGDP01151 11.05
Mozabite:HGDP01266 8.75
Cypriot:Cyprus2AJ19 6.65
Mozabite:HGDP01275 2.65
Cypriot:CYP19 2.35

Total Bergamo Italians: ~25%
Total Cypriot: ~64%
Total Mozabite: ~11%
Total S. French: disappeared again

I think it is quite clear, that the further you go north from historically Greek speaking Italy, the lower the amount of Italian admixture among Western Jews, and the more northern Italians almost always overlap with the S. French component.

Also, keep in mind that adding Italian to the admixture didn't change the distance dramatically - as one would expect if indeed, Italians were the S. European admixture among Western Jews.


*Another explanation for why even as far north as Bergamo Italians, Cypriots still lose around ~8%, while the distance is still small, is that perhaps there is a very small admixture with Italians, probably from Roman converts. After all, we all agree that before Western Jews diverged to Sephardim and Ashkenazim, they moved through Italy.
As I said before, I wouldn't reject the notion of Latin Romans conversion, or maybe even S. Italian conversion, but it must be very minimal - I think this exercise pretty much shown that Greek component in Western Jews is much more dominant than any Italian one.

Targum
06-11-2018, 02:32 PM
How Romaniote ancestrally many Jews from Turkey and Bulgaria today really are is difficult to gauge. For a good couple of centuries in Istanbul, a census of Jews there suggested it was 50/50 Sephardic and Romaniote. There was initially resistance to intermarry, but it seems like they were eventually assimilated into the Sephardic population. It was the same story for Greece, however I think more Yevanic-speaking Jews survived in Greece up until sadly the holocaust. Nazi's literally genocided the Yevanic-speaking Jews. You would be lucky to maybe find a minority today in Greece and maybe New York. Based on the Azmon study, Greek Jews (pretty sure they were Sephardic) were closest to Turkish Jews which isn't exactly surprising.

From memory I THINK i did have a gedmatch ID of someone who claimed their mother was full Romaniote partially from Greece and Turkey. I'll have to try fish from it, but I remembered her results were the same as Sephardic Jews, the same way Italqim's basically are.

Those sample of Italian Jews, how do we know they are Italqim? If they are from the Atzmon study they just come from the 'Italian Jewish community of Rome'. They could very well have absorbed a lot of Sephardic emigrees.

Atzmon study below.

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3032072/pdf/main.pdf

In addition to the dominant Sefaradi community which developed from Spanish Jewish exiles immediately pre- and post- Expulsion, Istanbul had the original Romaniote Anatolian original Jewish community. These were absorbed into the Sefaradim, and there is little contemporary Romaniote presence tod(even in places where there still Romaniote descendants, the Holocaust decimated them to near extinction -Gabrielle Carteris the U.S. actress is a Romaniote descendant). Few are aware that Istanbul's Ashkenazi community, which still exists with its own synagogue, albeit largely merged with the Sefaradi community, pre-dates the Sefaradim in Turkey by about 30 some years!So after the Romaniotes, Ashkenazim are the second oldest qehillah (organized congregation) in Istanbul. There also was a Karaite community at one time, and while I know there are individuals who claim Karaite ancestry, I am not sure if there is still a Karaite qehillah any more. Perhaps one of the Turkish Jews on here knows.

Erikl86
06-11-2018, 02:57 PM
Also, check out using population averages + showing single distances:

Sicilians vs. Cyprtions:

[1] "1. CLOSEST SINGLE ITEM DISTANCES"
Sicilian_East Cypriot Sicilian_West French_South Mozabite
0.01482157 0.01519200 0.01557596 0.04513900 0.06640698


[1] "distance%=0.7451 / distance=0.007451"

Sephardic_Jew

Cypriot 43.9
Sicilian_East 41.9
Mozabite 11.6
Sicilian_West 2.7
French_South 0.0



S. Italians vs. Cypriots:

[1] "1. CLOSEST SINGLE ITEM DISTANCES"
Cypriot Italian_South French_South Mozabite
0.01519200 0.01567970 0.04513900 0.06640698


[1] "distance%=0.857 / distance=0.00857"

Sephardic_Jew

Cypriot 51.65
Italian_South 34.25
Mozabite 11.65
French_South 2.45



Abruzzo Italians vs. Cypriots:

[1] "1. CLOSEST SINGLE ITEM DISTANCES"
Cypriot Italian_Abruzzo French_South Mozabite
0.01519200 0.01608933 0.04513900 0.06640698

[1] "distance%=0.83 / distance=0.0083"

Sephardic_Jew

Cypriot 50.2
Italian_Abruzzo 38.3
Mozabite 11.5
French_South 0.0


Tuscans vs. Cypriots:

[1] "1. CLOSEST SINGLE ITEM DISTANCES"
Cypriot Italian_Tuscan French_South Mozabite
0.01519200 0.02559329 0.04513900 0.06640698

[1] "distance%=0.9262 / distance=0.009262"

Sephardic_Jew

Cypriot 76.2
French_South 12.7
Mozabite 11.2
Italian_Tuscan 0.0

^^^ look how using average, Italian Tuscan actually disappears, rather than S. French.


Bergamo Italians vs. Cypriots:

[1] "1. CLOSEST SINGLE ITEM DISTANCES"
Cypriot Italian_Bergamo French_South Mozabite
0.01519200 0.02855601 0.04513900 0.06640698

[1] "distance%=0.8168 / distance=0.008168"

Sephardic_Jew

Cypriot 65.5
Italian_Bergamo 22.4
Mozabite 12.1
French_South 0.0

^^^ S. French disappears, while Cypriot also loses a bit. Check out history of Bergamo in Italy:


After a short period under the House of Malatesta starting from 1407, Bergamo was ceded in 1428 by the Duchy of Milan to the Republic of Venice in the context of the Wars in Lombardy and the aftermath of the 1427 Battle of Maclodio.
https://en.wikipedia.org/wiki/Bergamo

And as for Greeks in the territories of the Republic of Venice:

The migration of Byzantine Greek scholars and other emigres from Byzantium during the decline of the Byzantine empire (1203–1453) and mainly after the fall of Constantinople in 1453 until the 16th century, is considered by modern scholars as crucial in the revival of Greek and Roman studies, arts and sciences, and subsequently in the development of Renaissance humanism.[8] These emigres were grammarians, humanists, poets, writers, printers, lecturers, musicians, astronomers, architects, academics, artists, scribes, philosophers, scientists, politicians and theologians.[9]

In the decades following the Ottoman conquest of Constantinople many Greeks began to settle in territories of the Republic of Venice, including in Venice itself. In 1479 there were between 4000 and 5000 Greek residents in Venice.[10] Moreover, it was one of the economically strongest Greek communities of that time outside the Ottoman Empire.[11] In November 1494 the Greeks in Venice asked permission and were permitted to found a Brotherhood of the Greek race,[12] a philanthropic and religious society which had its own committee and officers to represent the interests of the flourishing Greek community. This was the first official recognition of the legal status of the Greek colony by the Venetian authorities.[13] In 1539 the Greeks of Venice were permitted to begin building their own church, the San Giorgio dei Greci which still stands in the centre of Venice in the present day on the Rio dei Greci.[13]
https://en.wikipedia.org/wiki/Greeks_in_Italy#Medieval


THE GREEK COMMUNITIES IN VENICE AND SOUTHERN ITALY.

By about 1478 the Greek population of Venice stood at some four thousand people, mainly concentrated in the Castello area of the city. This rapidly growing presence prompted Cardinal Bessarion to remark that Venice was 'almost another Byzantium'. In 1470 the Greeks were given a wing of the church of San Biagio in which to worship in their own language. In 1514 they received permission to build a church of their own and this was completed, as San Giorgio dei Greci, in 1573 (Geanakoplos, Greek Scholars, 35-7, 60; Nicol, Byzantium and Venice, 415-17). Many of these people appear to have found employment connected with Venice's position as a naval and mercantile power. They provided rowers for Venetian galleys, and carpenters for the Arsenal or shipyard. Between 1400 and 1442 a dynasty of Greek shipwrights dominated the Arsenal, designing galleys for both trade and war. Others, however, worked as tailors and gold wire drawers, or joined the Stradioti, a regiment in Venetian service recruited entirely from Greeks. An exception was Anna Notaras, a Byzantine noblewoman who had come to Venice before 1453 and who died there at an advanced age in 1507. Possessed of immense wealth, Notaras was financially independent and able to support many of her fellow Greeks in her household (Harris, Greek Emigres, 85-6, 180-1, 203; Nicol, Byzantium and Venice, 415-16).
https://www.arlima.net/the-orb/encyclop/late/laterbyz/harris-ren.html


So, it seems indeed - there was a late settlement of Greeks in N. Italy, which might explain why Cypriot suddenly loses percentage.

Except for East Sicilians, in closest single item distance all Italians fall below Cypriot average.

K33
06-11-2018, 03:01 PM
Sicilian_East
Bell_Beaker_Germany 0.222±0.077
England_Roman_outlier 0.210±0.134
Mycenaean 0.567±0.163
P-value 0.504442682
chisq 12.285
Full outputThat's interesting East Sicilians use the England Roman outlier (I assume this is 3DRIF-26), which was described in the original findings as closest to modern Saudis genetically but through teeth isotopes was most likely pegged as from the Sinai.

But we have a member here who uploaded Egyptian Copt genomes to Gedmatch, and 3DRIF-26 is quite clearly an Egyptian from Roman Egypt-- modern Muslim Egyptians have ~10% extra SSA which was throwing off their affinity to 3DRIF-26 hugely, but once the Copts got uploaded they were by far the best modern match.

This might imply that East Sicilians have actual Egyptian ancestry, but more likely is they have Carthaginian ancestry IMO. Carthaginians were probably midway between modern Levantines and Berbers on PCA, as are Roman-era and Coptic Egyptians.

Claudio
06-11-2018, 03:03 PM
It could make sense to me if we are to believe that Jews mixed with Greek-admxied people IN ITALY but there is nothing to suggest the admixture happened within the borders of modern Greece. As they moved up Italy they would have encountered people who were Greek-speaking. But I absolutely do not believe the admixture occurred while the Jews were still in the Levant.[/QUOTE]

I kind of agree with you and disagree.
I believe ‘some’ of the Greek Admixture was picked up in the Levant as proposed by Eric but I can’t help but agree with you that ‘most’ of the Greek was picked up within italy from centuries of living in heavily Greek Admixed areas of Italy like Southern Italy and Sicily.

I’m pretty sure I read that when Julius Caesar died the Jewish community of Rome mourned him as he did much for the Jewish community for instance allowing the Jewish communities allocation of grain to be organized around religious holidays where as later Emperors would sometimes victimized the Jewish community by allocating grain collection to clash with religious matters.
Did the majority of later Roman jews convert to Christianity?
Are the ancestors of western Jews perhaps Greko/Southern Italians Jews who moved up through Italy much later?

Erikl86
06-11-2018, 03:19 PM
Did the majority of later Roman jews convert to Christianity?


Yes, most Hellenistic Jewish communities were a fertile ground for early Christianity. One of the reasons why Early Church dropped circumcision is that they knew it'd buy them a lot of fans among Hellenistic Jews, which had a funny community of half-Jews, or "god fearer" Greeks - largely Greek men that didn't fully become Jews because of deterrence from circumcision. Other than the fact that it was extremely painful for a grown man to do it, it was also considered an abomination in Greco-Roman culture.

So most Hellenistic Jewish communities became the first Roman Christians. Also, after Jews became a hateful people - around 135 AD - it gave the culturally assimilated Hellenistic Jews a way out, because back then Christianity was still looked at as a type of Judaism, but not the same as those rebellious Judean Jews.




Are the ancestors of western Jews perhaps Greko/Southern Italians Jews who moved up through Italy much later?


I wouldn't reject at all the possibility that some Italian Greeks, or even Latin Romans, converted as there is documentation for that as well. All I'm saying, is that statistically, there were far more Greek Jews and Greek converts in the 1st century and Italian ones. So while perhaps there was some addition from Italy, which would also help increase the genetic affinity of modern Jews to Italians, I believe it was minor one - or we would have seen Tuscans dominant over Cypriots, or at least sharing closer to the 50/50 admixture we get when using Sicilians, or West Sicilians as close to Western Jews as East Sicilians.

It's clear that there is correlation between Greek ancestry and affinity to Western Jews among the different Italian groups.

kingjohn
06-11-2018, 03:25 PM
k36 nmonte of me remind me of lukasz work


With K36 oracle with modern populations, without Italians and Sicilians:
[1] "distance%=7.9383 / distance=0.079383"
Adam

GR_Dodecanes 53.45 {aegean stuff}
Spain_North 13.70 {catalans}
Samaritan 11.55 {levant}
GR_Cyclades 10.55 {aegean }
North_Africa 6.10 {moors}
PL_Sudovia 4.65 [slavs }

Now even if I put Sicilians and Italians, the most dominant component would be Greek:
[1] "distance%=7.4607 / distance=0.074607"
Adam

GR_Dodecanes 25.65 [aegean stuff}
Sicily_Messina 24.55{ aegean stuff}
IT_Lazio 19.40 {romans}
Samaritan 13.00 {levant honey }
Spain_North 7.40 {catalans}
PL_Sudovia 6.65 {slavs}
North_Africa 3.35 {moors}
And Messina and Lazio in Italy are both highly Greek in origins - see how closeness between Italians and Spanish lower the amount of Spanish admixture you have.
Also, Sicilians lower the amount of N. African, but actually the fact that we've used Sicilians to model you lowers the amount of Slavic admixture (because the Slavic influence in Greeks is absent in Italians).
When I used Cypriots vs. Tuscans, this is what you got (very similar to previous results in terms of distance):

[1] "distance%=7.4211 / distance=0.074211"
Adam
Cyprus 63.45 {aegean+levantine}
Spain_North 20.45 catalans { personally i prefer real madrid though}
Rusyn_NE_Slovakia 9.35 slavs :D
North_Africa 6.75 {moors}
France 0.00
Ukrainian_West 0.00
PL_Sudovia 0.00
PL_Wielkopolska 0.00
PL_SE_Carpathia 0.00
PL_Mazovia 0.00
PL_South_Poland 0.00
PL_Upper_Silesia 0.00
PL_Podlasie_East_Mazovia 0.00
IT_Tuscany 0.00
Again, proves that the reason why Sicilians got high percentage is because they have a lot of Greek ancestry.
I've even used Druze, instead of Samaritans, to see how would it go:
[1] "distance%=7.6403 / distance=0.076403"
Adam

Cyprus 63.0
Spain_North 20.5
Rusyn_NE_Slovakia 9.4
North_Africa 6.8
Druze 0.3
France 0.0
Ukrainian_West 0.0
PL_Sudovia 0.0
PL_Wielkopolska 0.0
PL_SE_Carpathia 0.0
PL_Mazovia 0.0
PL_South_Poland 0.0
PL_Upper_Silesia 0.0
PL_Podlasie_East_Mazovia 0.0
IT_Tuscany 0.0
And finally, a control group - of just sephardic average vs ashkenazi average (so you'd know if those ~7% distances are accurate enough):
[1] "distance%=6.718 / distance=0.06718"
Adam

eolien
06-11-2018, 03:28 PM
I don't know much about the Karaite community at present but I wonder how the Ashkenazi community is older that the Sefardic one in Istanbul. Contrary what you say, the origin and synagogues of the Ashkenazic community are very recent (about the last 100-150 years) and they did not merge with the Sefardic institutions. So even if you can prove the existence of an Ashkenazic community in Istanbul, it clearly did not survive.


In addition to the dominant Sefaradi community which developed from Spanish Jewish exiles immediately pre- and post- Expulsion, Istanbul had the original Romaniote Anatolian original Jewish community. These were absorbed into the Sefaradim, and there is little contemporary Romaniote presence tod(even in places where there still Romaniote descendants, the Holocaust decimated them to near extinction -Gabrielle Carteris the U.S. actress is a Romaniote descendant). Few are aware that Istanbul's Ashkenazi community, which still exists with its own synagogue, albeit largely merged with the Sefaradi community, pre-dates the Sefaradim in Turkey by about 30 some years!So after the Romaniotes, Ashkenazim are the second oldest qehillah (organized congregation) in Istanbul. There also was a Karaite community at one time, and while I know there are individuals who claim Karaite ancestry, I am not sure if there is still a Karaite qehillah any more. Perhaps one of the Turkish Jews on here knows.

Claudio
06-11-2018, 03:36 PM
Well we can explain it for the fact that Bulgarian and Turkish Jews are mainly Sephardic in origins, while Italkim have never been in Spain.



I agree with all of those assumptions and assume so as well.





No, I have no idea when will it be published.



Yes, but none of them are Romans from Rome. We need Roman-time Italian samples (hopefully not S. Italians as those would have been very close to Greeks already by Roman times).



Well Natufian is Neolithic Levantine, which is different than Bronze Age Levantine, mainly as a result of CHL Iranian migrations that happened 6000 years ago. But yes, it would make sense that Bronze Age Anatolian and Aegean would have very little Neolithic Levantine in them.

I’ve noticed on 23andMe Italian seems to peak in Abruzzo.
Bit off Topic but how do Abruzzo Italians come up and compare Component wise?
Intermediate between Tuscans and Southern Italians?

Targum
06-11-2018, 03:38 PM
No; it is not well known but in Istanbul proper the Ashkenazi qehillah pre-dated the Sefaradim by 30-some years. Below is an excerpt from the Wikipedia article on Istanbul Jewish history:

"The number of native Jews was soon bolstered by small groups of Ashkenazi Jews that immigrated to the Ottoman Empire between 1421 and 1453.[13] Among these new Ashkenazi immigrants was Rabbi Yitzhak Sarfati, a German-born Jew of French descent[15] (Hebrew: צרפתי – Sarfati, meaning: "French"), who became the Chief Rabbi of Edirne and wrote a letter inviting the European Jewry to settle in the Ottoman Empire, in which he stated that: "Turkey is a land wherein nothing is lacking" and asking: "Is it not better for you to live under Muslims than under Christians?"[15][16]..."