PDA

View Full Version : Major Jewish Y-lineages



Pages : [1] 2 3

hartaisarlag
07-29-2019, 04:22 PM
I agree with the concern in the Big Thread that we've spent too much time interpreting autosomal DNA-based models, and not enough making judgments based on clear phylogeny (which has gotten much more precise over the last few years). Drawing on a number of sources (JewishDNA.net, YTree, FTDNA's Haplotree, and FTDNA projects), I decided to put together brief, "just the facts" summaries on each of the 10 apparently largest Ashkenazi lineages, all with YFull-estimated TMRCAs between 400 CE and 900 CE. By Wim Penninx's original estimate from a few years ago, these make up about 55% of all Ashkenazi patrilineages.

For the most part I'll leave out my interpretations; I'm interested in yours (and in any blanks you'd be able to fill in). Suffice it to say, some branches have interesting internal structure and some don't; some have a clear link to non-Ashkenazi Jews, and some don't; some have a sibling branch within the historical timescale, and some don't; some appear Levantine, some European, and some unclear in origin.

Might be best to break this up, and post one by one, in descending order of size. Starting with a notorious one:

R1a-Z93-Y2619

Frequency (Penninx): 8.6%
TMRCA (YFull): 700 CE
TMRCA (Behar 2017): 250 CE

Main branches:
R-BY29826 (TMRCA: Not reported)
R-FGC18222 (TMRCA: 850 CE)
- Includes Iraqi Jew (in a downstream clade with Belarusian and Polish Jews with TMRCA: 1725 CE) and Algerian Jew (also in downstream clade)
R-Y2630 (Main branch; TMRCA: 950 CE)
- Includes Turkish Jews (in multiple downstream clades)

Parent branch:
R-CTS6 (TMRCA: 900 BCE)
- Basal individual: Persian from Kerman
- Haplotree identifies possible basal Egyptian, Colombian
- Found in a Scythian from ca. 1050 CE
Sibling branches:
R-Y37891 (TMRCA: 650 CE)
- Includes Persian from Kerman, Iranian Azerbaijani, Iraqi Yazidi
R-CTS7297 (TMRCA: 900 BCE)
- Includes Armenian, Iranian Azerbaijani, Brazilian, and Spaniards

Grandparent branch:
R-F1345 (TMRCA: 1900 BCE)
Uncle branches:
R-YP5484 (TMRCA: 1050 CE)
- Includes Galilee Palestinians (Mamluk tradition?)
R-F2935 (TMRCA: 1700 BCE)
- Includes Black Sea Turk, Balkars, Englishman, Scot, Spaniard, and Ukrainian

hartaisarlag
07-29-2019, 04:23 PM
Next:

J2a-M92-L556

Frequency (Penninx): 7.6%
TMRCA (YFull): 850 CE

Main branches:
J-Y9005 (TMRCA: 1250 CE)
- Includes early-diverging downstream subclade with Venetian and Bulgarian from Ruse. Jewish or not?
J-Y11782 (TMRCA: 1050 CE)
- Splits into Y22280 (TMRCA: 1150 CE) and Y13373 (TMRCA: 1500 CE); former has no German Jews and a Puerto Rican in a downstream branch, but latter has plenty of German Jews

Note:
Haplotree identifies possible basal Italians, Honduran, Mexican, Chilean, Pakistani.

Parent branch:
J-Y20051 (TMRCA: 4200 BCE)
Sibling branch:
J-S19243 (TMRCA: 1250 CE)
- All Saudis

Grandparent branch:
J-Y20492 (TMRCA: 4200 BCE)
- Basal individual: Sardinian
Uncle branch:
J-ZS6172 (TMRCA: 900 BCE)
- Western Turk, Armenian, Calabrian

hartaisarlag
07-29-2019, 04:24 PM
Next, for a very interesting one, still poorly-characterized:

G2b-M377-Y12975

Frequency (Penninx): 7.6%
TMRCA (YFull): 900 CE

Main branches:
G-Y14600 (TMRCA: 950 CE)
G-Y15861 (TMRCA: 1050 CE)

Parent branch:
G-Y12297 (TMRCA: 2700 BCE)
Sibling branch:
G-M3124 (TMRCA: 1250 CE)
- Sicilian, Afghan Pashtuns, Punjabi

Grandparent branch:
G-M377 (TMRCA: 6500 BCE)
- Lebanese, Syrians, Armenians, Tajiks, Neapolitan, Sicilian

hartaisarlag
07-29-2019, 04:51 PM
Including a downstream pan-North Mediterranean branch not reported in YFull:

Q1b-L245-Y2200

Frequency (Penninx): 6.4%
TMRCA (YFull): 450 CE

Main branches:
Q-YP1003 (TMRCA: 600 CE)
- According to Gurianov 2015, includes Chinese individual
Q-Y2197 (TMRCA: 700 CE)

Notes:
- Structure reported differently in YTree vs. Haplotree vs. GeneticHomeland.
- YFull identifies basal Scot; Haplotree additionally identifies basal Dane and Englishman.
- Haplotree identifies a Brazilian, Italian, Greek, and Portuguese under Q-BZ72, a sibling to Q-Y2197, under Q-FGC1933, which is the actual sibling to Q-YP1003.

Parent branch:
Q-Y2225 (TMRCA: 500 BCE)
Sibling branch:
Q-FGC10659 (TMRCA: 1875 CE)
- Italians

Grandparent branch:
Q-Y2209 (TMRCA: 1400 BCE)
- Basal individual: Jordanian, Haplotree adds Lebanese, Omani, German (Jew?)
Uncle branches:
Q-FGC2020
- Turk (or Armenian from Turkey?)
Q-BZ1
- Includes Italian
Q-BZ3000 (TMRCA: 750 CE)
- Includes Montenegrin, Croat
Q-YP730 (TMRCA: 900 BCE)
- Includes Chinese, Dutch, German

hartaisarlag
07-29-2019, 04:53 PM
Near and dear to my heart:

E1b-L791-Y6923

Frequency (Penninx): 5.6%
TMRCA (YFull): 400 CE, main branch: 700 CE

Main branches:
E-Y102667 (TMRCA: 400 CE)
- Splits into Y102667* (Libyan Jew) and Y99093 (TMRCA: 650 CE, Turkish-Tunisian Jew and Puerto Rican)
- Most likely (Y6923 xY6938) includes Algerian (Jew?), Mexican, Peruvian
- Might include Ashkenazi (Behar et al 2017)
E-Y6938 (TMRCA: 700 CE)
- Primary split between Y133118 (TMRCA: 950 CE) and main clade Z36123 (TMRCA: 750 CE)
- Found in downstream clades: Swedes, a Scot, a Polish Tatar (?)

Parent branch:
E-Y6926 (TMRCA: ~1550 BCE)
Sibling branch:
E-Y6926*
- Includes Omanis and Emiratis from Al-Bulushi clan

Grandparent branch:
E-Y4971 (TMRCA: 3500 BCE)
- Haplotree identifies possible basal Emirati, Polish (Jew?), Russian (Jew?)
Uncle branches:
E-BY36592 (TMRCA: Not reported)
- Not well-characterized, Iraqi(s)
E-Y4972 (TMRCA: 3000 BCE)
- Swiss German, Romaniote Jews, Sardinian, Armenian, Russian, Ligurians, French, Nablus-area Palestinian, Kuwaitis, Emiratis, German (Jews?), Iraqi, Portuguese, Spaniard, Venezuelan

Note:
Possibly identified in a Bulgarian (FTDNA), and in now-inaccessible databases, a Brazilian and Tunisian (Jew?).

hartaisarlag
07-29-2019, 05:17 PM
Shoutout to the priests:

J1-ZS227-Y5400 (S12192)

Frequency (Penninx): 5%
TMRCA (YFull): 400 CE

Main branches:
J-Y31161 (TMRCA: 400 CE)
J-Y5399 (TMRCA: 950 CE)

Notes:
- Nomenclature different in YTree vs. Haplotree.
- Haplotree identifies possible basal Moroccan (Jew?), Algerian (Jew?), Ashkenazim. Includes Norwegian and Irish in downstream branches.

Parent branch:
J-Y3088 (Y18271) (TMRCA: 900 BCE)
- Includes Turkish (Jew?), Englishman, Sardinian, Peruvian, Greeks (Jewish?), Yemenite Jew, Salvadoran, Lebanese, Mexican, and Puerto Rican.
Sibling branches:
J-ZS2432 (TMRCA: 750 CE)
- Ashkenazim
J-ZS2458 (TMRCA: 800 BCE)
- Iraqi (Jew?)
J-FGC17491 (TMRCA: 250 CE)
- Includes Portuguese, Spaniards, Moroccan Jew, Mexican, Ashkenazim
J-ZS237 (TMRCA: Not reported)
- Italians

Grandparent branch:
J-S0075 (ZS227) (TMRCA: 3400 BCE)
- Haplotree identifies possible basal Iraqi (Jew?), Kuwaiti, French (Jew?), Ashkenazim; Portuguese at ZS222 (between ZS227 and Y18271). YFull identifies basal Rhineland German (Jew?).
- JewishDNA.net claims (based on STR?) Jews from Turkey, Italy, Iran, Syria, Lebanon, and Egypt.

hartaisarlag
07-29-2019, 05:19 PM
J2a-L210-Y15223 (Z30390)

Frequency (Penninx): 4%
TMRCA (YFull): 550 CE

Main branches:
J-BY194993 (TMRCA: 1250 CE)
J-Y15238 (TMRCA: 1100 CE)
- Includes several German Jews on YFull (3 Bavarian, 1 Rhineland)
J-Y24492 (TMRCA: 550 CE)
- Doesn’t include any German Jews on YFull

Notes:
- Nomenclature different in YTree vs. Haplotree.
- Haplotree identifies possible basal Mexican, Ashkenazim.

Parent branch:
J-Y15222 (TMRCA: 600 BCE)
- Includes basal Lombard, Basilicatan. Haplotree reports possible basal Frenchman and Irishman. No downstream subclades reported (other than Y15223).

Grandparent branch:
J-Z482 (TMRCA: 2900 BCE)
- Includes basal Ukrainian (Jew?); Haplotree reports possible basal Ashkenazim (or Eastern Europeans), Italians, Danes, Spaniard, Scot.
Uncle branch:
- J-Z478 (TMRCA: Not reported)
- Tuscans

hartaisarlag
07-29-2019, 05:21 PM
I like this one because it strongly hints at intra-Ashkenazi structure:

E1b-M84-Y14891

Frequency (Penninx): 4%
TMRCA (YFull): 800 CE

Main branches:
E-Z36149 (TMRCA: 1250 CE)
- Includes several German Jews (Bavarian, Rhineland, Baden-Wurttemberg), Alsatian Jews, Dutch Jew
E-Y16781 (TMRCA: 800 CE)
- Larger branch; includes much smaller proportion of German Jews, plus Venetian Jew Bassan downstream (Ashkenazi origin likely)

Parent branch:
E-Y14899 (TMRCA: 5500 BCE)
- Appears to include Syrian Jew, somewhere in structure
Sibling branches:
E-F1539 (TMRCA: 300 CE)
- Includes Kuwaiti, Emirati
E-Y14899* (BY63615?)
- Includes Palestinian Christian from Birzeit, Greek Cypriot with possible Pontic origin

Grandparent branch:
E-FGC18353 (TMRCA: 5500 BCE)
Uncle branch:
E-Y5435 (TMRCA: 4000 BCE)
- Includes Yemenis, Saudis, Qataris, Emirati, Iranian Kurd, Sardinians, Algerian, Bulgarian, Lebanese, Bahraini, Palestinian, Swede, Swiss, Turk, Iraqi, English, Italian, Austrian, Polish, Hungarian.
- Includes diverse, heavily but not exclusively European branch E-FGC18401 downstream, (TMRCA: 1100 BCE).

hartaisarlag
07-29-2019, 05:22 PM
J1-L817-L816

Frequency (Penninx): 3.3%
TMRCA (YFull): 850 CE

Main branches:
J-Y34527 (TMRCA: 1350 CE)
- Includes Mexicans from Nuevo León, Spaniard, German Jew
J-ZS2728 (TMRCA: 900 CE)
- Much larger branch; includes basal Spaniard, downstream Finn

Note:
Haplotree includes possible basal Greek (Jew?), Syrian (Jew?), Ashkenazim.

Parent branch:
J-L818 (TMRCA: 1900 BCE)
According to Haplotree, includes possible basal Turks (Kurds?), Italian, Syrian, Saudi, German (Jew?).

Grandparent branch:
J-L817 (TMRCA: 4600 BCE)
- YTree includes basal Sardinian
Uncle branch:
J-ZS5383 (TMRCA: Not reported)
- Missing from YTree
- Includes Yemenis, Saudis, Palestinian, Iraqi

hartaisarlag
07-29-2019, 05:22 PM
R1b-U106-Y5051

Frequency (Penninx): 2.9%
TMRCA (YFull): 500 CE

Main branches:
R-Y5051*
- Includes Portuguese, Mexican, Brazilian, Spaniard
R-FGC8591 (TMRCA: 850 CE)
- Downstream of R-FGC8564 (TMRCA: 500 CE)
R-A689 (TMRCA: 1200 CE)
- Downstream of R-FGC8564 (TMRCA: 500 CE)
- Includes Turkish Jew

Note:
Nomenclature different in YTree vs. Haplotree.

Parent branch:
R-Y3464 (FGC8587) (TMRCA: 500 CE)
- Includes basal Englishman; according to Haplotree, includes additional possible basal Englishman, Swede.

Grandparent branch:
R-Y3462 (TMRCA: 100 CE)
Sibling branch:
R-Y101927 (TMRCA: 1200 CE)
- Includes Englishmen, might also include German

Agamemnon
07-29-2019, 05:41 PM
Shoutout to the priests:

J1-ZS227-Y5400 (S12192)

Frequency (Penninx): 5%
TMRCA (YFull): 400 CE

Main branches:
J-Y31161 (TMRCA: 400 CE)
J-Y5399 (TMRCA: 950 CE)

Notes:
- Nomenclature different in YTree vs. Haplotree.
- Haplotree identifies possible basal Moroccan (Jew?), Algerian (Jew?), Ashkenazim. Includes Norwegian and Irish in downstream branches.

Parent branch:
J-Y3088 (Y18271) (TMRCA: 900 BCE)
- Includes Turkish (Jew?), Englishman, Sardinian, Peruvian, Greeks (Jewish?), Yemenite Jew, Salvadoran, Lebanese, Mexican, and Puerto Rican.
Sibling branches:
J-ZS2432 (TMRCA: 750 CE)
- Ashkenazim
J-ZS2458 (TMRCA: 800 BCE)
- Iraqi (Jew?)
J-FGC17491 (TMRCA: 250 CE)
- Includes Portuguese, Spaniards, Moroccan Jew, Mexican, Ashkenazim
J-ZS237 (TMRCA: Not reported)
- Italians

Grandparent branch:
J-S0075 (ZS227) (TMRCA: 3400 BCE)
- Haplotree identifies possible basal Iraqi (Jew?), Kuwaiti, French (Jew?), Ashkenazim; Portuguese at ZS222 (between ZS227 and Y18271). YFull identifies basal Rhineland German (Jew?).
- JewishDNA.net claims (based on STR?) Jews from Turkey, Italy, Iran, Syria, Lebanon, and Egypt.

It's not just S12192 that is tied to Kohanim though, but Z18271 as a whole. I'm ZS2432 for instance. The Algerian, Greek (except a handful of non-Jewish Pontians), Lebanese and Yemenite samples are all Kohanim as well.

hartaisarlag
07-29-2019, 05:48 PM
It's not just S12192 that is tied to Kohanim though, but Z18271 as a whole. I'm ZS2432 for instance. The Algerian, Greek (except a handful of non-Jewish Pontians), Lebanese and Yemenite samples are all Kohanim as well.

Thank you - this is what I came for.

Do you trust the TMRCA of 900 BCE? YFull is already behind on the structure of the branch, so I wouldn't be surprised if timing were off too.

Do you know if the Iraqi on the tree is Jewish?

AbdoNumen
07-29-2019, 06:05 PM
Thank you - this is what I came for.

Do you trust the TMRCA of 900 BCE? YFull is already behind on the structure of the branch, so I wouldn't be surprised if timing were off too.

Do you know if the Iraqi on the tree is Jewish?

Though the question isn't directed at me:

Behar at al. 2017 date Z18271 to ~570 BCE.

The Iraqi on the tree is Jewish, but not clear whether he has Cohen tradition (Mallick et al 2016).

vettor
07-29-2019, 06:07 PM
https://sites.google.com/site/levitedna/y-dna-haplogroups-of-ashkenazi-jews/haplogroup-t-y-dna-clusters-for-ashkenazi-jews

see other haplogroups on the left side of link

hartaisarlag
07-29-2019, 06:09 PM
double post

hartaisarlag
07-29-2019, 06:09 PM
https://sites.google.com/site/levitedna/y-dna-haplogroups-of-ashkenazi-jews/haplogroup-t-y-dna-clusters-for-ashkenazi-jews

see other haplogroups on the left side of link

I'm aware of this resource; for everything besides the R1a-Y2619 stuff, I prefer JewishDNA.net. Wexler was under the impression that most Ashkenazi R1b was Near Eastern, based on a big phylogenetic misinterpretation.

Agamemnon
07-29-2019, 06:10 PM
Thank you - this is what I came for.

Do you trust the TMRCA of 900 BCE? YFull is already behind on the structure of the branch, so I wouldn't be surprised if timing were off too.

Do you know if the Iraqi on the tree is Jewish?

I know there is one Iraqi who isn't Jewish, while the rest are. IIRC the non-Jewish Iraqi is negative for Z18271, so the Iraqi under ZS2458 is Jewish. The confusing thing is that they all seem to come from Anbar province. I do not know whether this individual has a tradition of priestly descent, I've seen nothing that suggests this so far.

Also, Z18290/FGC17491 contains at least one Cohen I know of from Lithuania (kit 25947). So its presence in Iberia is almost undoubtedly Jewish, and more to the point Kohanic. This clade is a good example of how Z18271 as a whole is associated with Jewish priesthood.

Finally, there is at least one Yemenite Muslim in there, namely Al Abdali from Wadi Yahr (kit 476158), this individual is Z18271 (xZS237, xS12192, xZ18290). This is very likely to be of recent Jewish origin as well, interestingly several of the Jewish tribes of Arabia claimed Kohanic descent, the presence of Z18271 gives a lot of weight to this claim. The Banu Qurayza were largely thought to be of Kohanic descent for instance.


Though the question isn't directed at me:

Behar at al. 2017 date Z18271 to ~570 BCE.

The Iraqi on the tree is Jewish, but not clear whether he has Cohen tradition (Mallick et al 2016).

I want to add that I agree with Behar et al.'s TMRCA estimate for the time being and expect the estimates on YFull to go down as future results are added to the tree.

vettor
07-29-2019, 06:34 PM
I'm aware of this resource; for everything besides the R1a-Y2619 stuff, I prefer JewishDNA.net. Wexler was under the impression that most Ashkenazi R1b was Near Eastern, based on a big phylogenetic misinterpretation.

I differ

for me and my T1a2 branch , there is only 4 samples under CTS8489 ( equal to CTS8862 ) , stating from Galicia Poland ( your link )...........but my link goes one step further, to SNP T-Pages00113 which is found in Galicians, Germans and Kurds

BTW , I am negative for Pages00113

AbdoNumen
07-29-2019, 06:34 PM
Not sure how widespread it is, but there's one eastern European Ashkenazi Levite lineage (E1b-M84-BY36976 on FTDNA) that share ancestry with Iraqi Arabs from Tikrit-Kirkuk (E-Y83770, TMRCA ~2000 ybp according to Yfull).
I share a common ancestor with them sometime in the Bronze Age (E-Z21421 or Y18353, ~2300 BCE according to Yfull) -- me being a Jew from the Caucasus, no Cohen or Levite tradition.

hartaisarlag
07-29-2019, 07:00 PM
Not sure how widespread it is, but there's one eastern European Ashkenazi Levite lineage (E1b-M84-BY36976 on FTDNA) that share ancestry with Iraqi Arabs from Tikrit-Kirkuk (E-Y83770, TMRCA ~2000 ybp according to Yfull).
I share a common ancestor with them sometime in the Bronze Age (E-Z21421 or Y18353, ~2300 BCE according to Yfull) -- me being a Jew from the Caucasus, no Cohen or Levite tradition.

Penninx only reports one (probably that individual), but it's likely more will eventually turn up.

Erikl86
07-29-2019, 08:19 PM
Excellent thread ! :)




R1a-Z93-Y2619

Frequency (Penninx): 8.6%
TMRCA (YFull): 700 CE
TMRCA (Behar 2017): 250 CE

Main branches:
R-BY29826 (TMRCA: Not reported)
R-FGC18222 (TMRCA: 850 CE)
- Includes Iraqi Jew (in a downstream clade with Belarusian and Polish Jews with TMRCA: 1725 CE) and Algerian Jew (also in downstream clade)
R-Y2630 (Main branch; TMRCA: 950 CE)
- Includes Turkish Jews (in multiple downstream clades)

Parent branch:
R-CTS6 (TMRCA: 900 BCE)
- Basal individual: Persian from Kerman

Which one is the Levite one? IMO, this entered to Western Jews following the Geonic period of scholars from Babylonia settling in Spain and Ashkenaz, teaching Talmud.

hartaisarlag
07-29-2019, 08:26 PM
Excellent thread ! :)



Which one is the Levite one? IMO, this entered to Western Jews following the Geonic period of scholars from Babylonia settling in Spain and Ashkenaz, teaching Talmud.

I believe all of Y2619, in its three major branches, is Levite, with a very small number of exceptions.

I'm doubtful about the Babylonian interpretation, because all of the non-Ashkenazi Jews in Y2619 are located way downstream. If anything, I think an Ashkenazi origin for those Mizrahi/Sephardi patrilineages is more parsimonious. The Iraqi Jew has a YFull TMRCA of 375 ybp with a Polish Jew, a Belarusian Jew, and 4 newly uploaded samples (all evidently from Behar 2017).

Erikl86
07-29-2019, 08:35 PM
I believe all of Y2619, in its three major branches, is Levite, with a very small number of exceptions.

I'm doubtful about the Babylonian interpretation, because all of the non-Ashkenazi Jews in Y2619 are located way downstream. If anything, I think an Ashkenazi origin for those Mizrahi/Sephardi patrilineages is more parsimonious. The Iraqi Jew has a YFull TMRCA of 375 ybp with a Polish Jew, a Belarusian Jew, and 4 newly uploaded samples (all evidently from Behar 2017).

I just saw that on YFull myself, and I agree - it makes more sense that an Ashkenazi Jew perhaps made its way to Iraq - perhaps even during Ottoman times.

Nive1526
07-29-2019, 08:48 PM
J2a-L210-Y15223 (Z30390)


Grandparent branch:
J-Z482 (TMRCA: 2900 BCE)
- Includes basal Ukrainian (Jew?); Haplotree reports possible basal Ashkenazim (or Eastern Europeans), Italians, Danes, Spaniard, Scot.


There are two Ukrainians with very likely Jewish surnames and background in the FTDNA J2 project. The branch is labeled as Z482 (xZ478,Z30390), I don't know if posting kit numbers is encouraged here.
Probably they are the same or related to the haplotree individuals.

hartaisarlag
07-29-2019, 09:34 PM
I just saw that on YFull myself, and I agree - it makes more sense that an Ashkenazi Jew perhaps made its way to Iraq - perhaps even during Ottoman times.

Strange, but these phylogenies so often are, at one level or another.

What I appreciate about them is how clarifying the structure can be. The difference between migration from one community to another, vs. early divergence, for example.

Agamemnon
07-29-2019, 10:36 PM
J2a-L210-Y15223 (Z30390)

Frequency (Penninx): 4%
TMRCA (YFull): 550 CE

Main branches:
J-BY194993 (TMRCA: 1250 CE)
J-Y15238 (TMRCA: 1100 CE)
- Includes several German Jews on YFull (3 Bavarian, 1 Rhineland)
J-Y24492 (TMRCA: 550 CE)
- Doesn’t include any German Jews on YFull

Notes:
- Nomenclature different in YTree vs. Haplotree.
- Haplotree identifies possible basal Mexican, Ashkenazim.

Parent branch:
J-Y15222 (TMRCA: 600 BCE)
- Includes basal Lombard, Basilicatan. Haplotree reports possible basal Frenchman and Irishman. No downstream subclades reported (other than Y15223).

Grandparent branch:
J-Z482 (TMRCA: 2900 BCE)
- Includes basal Ukrainian (Jew?); Haplotree reports possible basal Ashkenazim (or Eastern Europeans), Italians, Danes, Spaniard, Scot.
Uncle branch:
- J-Z478 (TMRCA: Not reported)
- Tuscans

While Y15222 is indeed found in Italians, it is very likely to be Jewish in origin. L210 has been found in Lebanon, Syria and Armenia. Moreover, L210 is a branch of Z467, which was found in two of the MBA Anatolians (labeled Hittite but more likely to be Hattian in origin along the paternal line), while these samples are not L210 (they are Z6271) this helps us anchor L210 in a West Asian context even further. Coupled with the Z489* individual from Armenia, this is one of the branches that is likely to have arrived with KA-derived Khirbet Kerak Ware in the Levant during the EB II to EB III period.

L210 does have a complex history, however the small description above does not do it justice and might even give the (IMO) false impression that this lineage is bound to represent some form of European introgression, while in fact it is more likely to be Levantine in origin.


I believe all of Y2619, in its three major branches, is Levite, with a very small number of exceptions.

I'm doubtful about the Babylonian interpretation, because all of the non-Ashkenazi Jews in Y2619 are located way downstream. If anything, I think an Ashkenazi origin for those Mizrahi/Sephardi patrilineages is more parsimonious. The Iraqi Jew has a YFull TMRCA of 375 ybp with a Polish Jew, a Belarusian Jew, and 4 newly uploaded samples (all evidently from Behar 2017).

Just upstream from CTS6 we have F1345, one of the basal clades for this lineage is YP5484 which comprises two Palestinian Muslims from Tzfat/Safed, one of whom is a member on this forum (Mamluk). While Jamal's (Mamluk) family has a tradition of descent from the Mamluks (and more specifically from the Cumans) in-keeping with his family name, the TMRCA of F1345 makes an arrival with the Mitanni quite plausible, it would be wortwhile to keep this in mind. R1a-CTS6 could potentially be much older in the Levant than the scenario envisioned here.

artemv
07-29-2019, 11:49 PM
Expected to see at least some branches with TMRCA BCE.
We have discussed earlier that most Roman Empire era Jews later converted to Christianity.
TMRCA data well confirms it.

hartaisarlag
07-30-2019, 12:14 AM
Expected to see at least some branches with TMRCA BCE.
We have discussed earlier that most Roman Empire era Jews later converted to Christianity.
TMRCA data well confirms it.

All subject to change. With the exception of the Kohanic branch Z18271, I don't know of major ones with BCE TMRCA's.

Erikl86
07-30-2019, 07:14 AM
Including a downstream pan-North Mediterranean branch not reported in YFull:

Q1b-L245-Y2200

Frequency (Penninx): 6.4%
TMRCA (YFull): 450 CE

Main branches:
Q-YP1003 (TMRCA: 600 CE)
- According to Gurianov 2015, includes Chinese individual
Q-Y2197 (TMRCA: 700 CE)

Notes:
- Structure reported differently in YTree vs. Haplotree vs. GeneticHomeland.
- YFull identifies basal Scot; Haplotree additionally identifies basal Dane and Englishman.
- Haplotree identifies a Brazilian, Italian, Greek, and Portuguese under Q-BZ72, a sibling to Q-Y2197, under Q-FGC1933, which is the actual sibling to Q-YP1003.

Parent branch:
Q-Y2225 (TMRCA: 500 BCE)
Sibling branch:
Q-FGC10659 (TMRCA: 1875 CE)
- Italians

Grandparent branch:
Q-Y2209 (TMRCA: 1400 BCE)
- Basal individual: Jordanian, Haplotree adds Lebanese, Omani, German (Jew?)
Uncle branches:
Q-FGC2020
- Turk (or Armenian from Turkey?)
Q-BZ1
- Includes Italian
Q-BZ3000 (TMRCA: 750 CE)
- Includes Montenegrin, Croat
Q-YP730 (TMRCA: 900 BCE)
- Includes Chinese, Dutch, German

Here's a complete branch of Jewish Q from the Jewish Q FTDNA project (Ashkenazi, Mizrahi and Yemenite):

http://i64.tinypic.com/oks978.png

And here's the tree from Gurianov et al. (2014):

http://i68.tinypic.com/mj9zpv.png

AJ1 and AJ2 are of course both Ashkenazi Jews.

https://www.researchgate.net/publication/274510291_Phylogenetic_Structure_of_Q-M378_Subclade_Based_On_Full_Y-Chromosome_Sequencing

hartaisarlag
07-30-2019, 03:48 PM
Here's a complete branch of Jewish Q from the Jewish Q FTDNA project (Ashkenazi, Mizrahi and Yemenite):

http://i64.tinypic.com/oks978.png

And here's the tree from Gurianov et al. (2014):

http://i68.tinypic.com/mj9zpv.png

AJ1 and AJ2 are of course both Ashkenazi Jews.

https://www.researchgate.net/publication/274510291_Phylogenetic_Structure_of_Q-M378_Subclade_Based_On_Full_Y-Chromosome_Sequencing

I'm going off the most recent FTDNA Haplotree, which seems most current and detailed. AFAIK nobody's written about BZ72.

hartaisarlag
07-30-2019, 03:58 PM
While Y15222 is indeed found in Italians, it is very likely to be Jewish in origin. L210 has been found in Lebanon, Syria and Armenia. Moreover, L210 is a branch of Z467, which was found in two of the MBA Anatolians (labeled Hittite but more likely to be Hattian in origin along the paternal line), while these samples are not L210 (they are Z6271) this helps us anchor L210 in a West Asian context even further. Coupled with the Z489* individual from Armenia, this is one of the branches that is likely to have arrived with KA-derived Khirbet Kerak Ware in the Levant during the EB II to EB III period.

L210 does have a complex history, however the small description above does not do it justice and might even give the (IMO) false impression that this lineage is bound to represent some form of European introgression, while in fact it is more likely to be Levantine in origin.

Thanks for this perspective. Seems too early to say, though. BA Anatolia > Italy > Ashkenazim seems plausible as well. Either way, I'll make sure to keep an eye out for other hints of a relatively recent Near Eastern connection.


Just upstream from CTS6 we have F1345, one of the basal clades for this lineage is YP5484 which comprises two Palestinian Muslims from Tzfat/Safed, one of whom is a member on this forum (Mamluk). While Jamal's (Mamluk) family has a tradition of descent from the Mamluks (and more specifically from the Cumans) in-keeping with his family name, the TMRCA of F1345 makes an arrival with the Mitanni quite plausible, it would be wortwhile to keep this in mind. R1a-CTS6 could potentially be much older in the Levant than the scenario envisioned here.

Right. First, is Mamluk one of the ERS samples, or his he the YFull individual? If the latter's true, then we have 3 Palestinians in F1345. (I guess two of them appear to be immediately related, so my point doesn't make much of a difference.) A TMRCA of 1050 CE seems a little early for a common Mamluk origin, but who knows. Either way, CTS6 seems to have a distinctly Iranic cast—especially given that it got into Yazidis, who I'd assume were not a very cosmopolitan population.

Fair to hypothesize that the clade's Iberians are of Alan origin? Because of how far upstream they are from the Ashkenazi branch, it seems unlikely that it's a Sephardic thing.

JoeyP37
07-30-2019, 04:03 PM
Also there is the Jewish subclade of European R1a, YP1013, a descendant of the Eastern European YP417, cousin to my YP445, as both are descended from the L1029 subclade of M458. Probably entered the Ashkenazic community in Poland, Belarus, or western Russia.

hartaisarlag
07-30-2019, 04:21 PM
Also there is the Jewish subclade of European R1a, YP1013, a descendant of the Eastern European YP417, cousin to my YP445, as both are descended from the L1029 subclade of M458. Probably entered the Ashkenazic community in Poland, Belarus, or western Russia.

Sure looks Slavic—I wouldn't be able to guess anything more particular than that. If we're to trust Penninx's numbers, it's found at a rate of about 0.5% in Ashkenazim.

Erikl86
07-30-2019, 04:42 PM
Sure looks Slavic—I wouldn't be able to guess anything more particular than that. If we're to trust Penninx's numbers, it's found at a rate of about 0.5% in Ashkenazim.

Which might actually be the fabled gene flow alleged to originated from pogroms. Rate of 0.5% could actually come from rape (which only rarely ends up in pregnancies).

hartaisarlag
07-30-2019, 04:58 PM
Could be, but no reason to rule out male conversion, rare as it appears to have been.

Looks like the two Ashkenazim in this subclade on YFull have a TMRCA of ca. 750 CE, for what it's worth.

Agamemnon
07-30-2019, 05:24 PM
Right. First, is Mamluk one of the ERS samples, or his he the YFull individual? If the latter's true, then we have 3 Palestinians in F1345. (I guess two of them appear to be immediately related, so my point doesn't make much of a difference.) A TMRCA of 1050 CE seems a little early for a common Mamluk origin, but who knows. Either way, CTS6 seems to have a distinctly Iranic cast—especially given that it got into Yazidis, who I'd assume were not a very cosmopolitan population.

Fair to hypothesize that the clade's Iberians are of Alan origin? Because of how far upstream they are from the Ashkenazi branch, it seems unlikely that it's a Sephardic thing.

No, he's one of the ERS samples. That's also what I thought, the TMRCA doesn't really fit with an arrival via the Mamluks. Hence my assumption that this branch might actually have been in the Levant since the MBA. That being said you are absolutely right to note the closer Iranian associations, but I for one would not be too quick to assign CTS6 specifically to Iranian populations, one of the often overlooked facts in the spread of Indo-Iranian is that Old Indic predated Iranian languages in the Iranian plateau. Thomas Burrow for instance proposed a wave of Indo-Aryan settlement in the Iranian plateau on the basis of linguistics and comparative mythology (one of the arguments relies on the presence of an Indo-Aryan substrate in Avestan), I could go into details but I'm not sure this is the right place for such a discussion.

What I am trying to say is that the presence of F1345 in West Iranian-speaking groups does not necessarily mean that the lineage itself arrived with the Proto-Iranians, it could easily be older and possibly even related to the initial wave of Indo-Iranian speakers (which was Old Indic) of which the Mitanni (and possibly the Kassites to some extent) were the westernmost outcome. Naturally, it would have re-expanded with later Iranian migrations with groups such as the Alans and the Sarmatians, which is IMO the most parsimonious explanation for the presence of F2935 in Western Europe.

By the way much of what I just wrote is bound to count for Q-L245.

josh w.
07-30-2019, 05:27 PM
All subject to change. With the exception of the Kohanic branch Z18271, I don't know of major ones with BCE TMRCA's. once identified

I don't know about TMCA, but I belong to a minor J2a subclade with origins in the Zagros mountains. The subclade is upstream of J2a4 (L26, L27) the more common Jewish pattern. The closest subclade is an upstream group, the Svans, with origins in Armenia. My subclade has an additional mutation. Ftdna J2a Project once described the subclade as Jewish but there are a couple of non Jewish Levantine members. My hunch is that the line migrated west on the Euphrates to Aleppo.

Erikl86
07-30-2019, 05:33 PM
Could be, but no reason to rule out male conversion, rare as it appears to have been.

Looks like the two Ashkenazim in this subclade on YFull have a TMRCA of ca. 750 CE, for what it's worth.

Sure, I can't rule out conversion, especially with such a rare occurrence, but IMO male conversion, is almost unheard of that late (post Roman period) in history, and as evident from the gender bias of Eastern European in favor of maternal lineages among Ashkenazi Jews, I find it dubious that it indeed entered via male conversion. Can't rule that out, but still, less likely.

Btw, a general warning about any assumptions we make regarding the occurrences of uniparentals in Ashkenazi Jews: always keep in mind the genetic bottleneck.

hartaisarlag
07-30-2019, 06:57 PM
once identified

I don't know about TMCA, but I belong to a minor J2a subclade with origins in the Zagros mountains. The subclade is upstream of J2a4 (L26, L27) the more common Jewish pattern. The closest subclade is an upstream group, the Svans, with origins in Armenia. My subclade has an additional mutation. Ftdna J2a Project once described the subclade as Jewish but there are a couple of non Jewish Levantine members. My hunch is that the line migrated west on the Euphrates to Aleppo.

What's the subclade?

artemv
07-30-2019, 07:16 PM
Sure, I can't rule out conversion, especially with such a rare occurrence, but IMO male conversion, is almost unheard of that late (post Roman period) in history, and as evident from the gender bias of Eastern European in favor of maternal lineages among Ashkenazi Jews, I find it dubious that it indeed entered via male conversion. Can't rule that out, but still, less likely.

Btw, a general warning about any assumptions we make regarding the occurrences of uniparentals in Ashkenazi Jews: always keep in mind the genetic bottleneck.

That's not the only case.
There are other Ashkenazi Y-chromosome lineages of European descent.
Like Y23115 for example.

There where periods of time, when Jewish communities in Christian Europe where rich and well established, it could happen that there where male Christians who wanted to convert.
But we now understand that both the convert and Jews, who accepted him, would prefer to keep it secret. So, no doubt there will be no historical records about such facts. We know about a few cases of Christians executed tor an attempt to convert to Judaism, but this is the case when we only know about those who failed to keep it secret.
That is true that more converts were female, but we should not overestimate the percentage. If let's say 20-25% of converts where male this also gives us a significant number.

hartaisarlag
07-30-2019, 07:22 PM
That's not the only case.
There are other Ashkenazi Y-chromosome lineages of European descent.
Like Y23115 for example.

There where periods of time, when Jewish communities in Christian Europe where rich and well established, it could happen that there where male Christians who wanted to convert.
But we now understand that both the convert and Jews, who accepted him, would prefer to keep it secret. So, no doubt there will be no historical records about such facts. We know about a few cases of Christians executed tor an attempt to convert to Judaism, but this is the case when we only know about those who failed to keep it secret.
That is true that more converts were female, but we should not overestimate the percentage. If let's say 20-25% of converts where male this also gives us a significant number.

I believe a Jewish friend of mine (well, Jewish paternal grandfather) has that haplogroup (AFAIK, it's the only attested Ashkenazi branch of I-M423)—it was a bit strange to see that result when he first got his 23andMe back. Originally Balkan, it would seem? Heavy Greek skew; TMRCA of all Greek branches before the entry of Slavs into Greece.

Erikl86
07-30-2019, 07:55 PM
That's not the only case.
There are other Ashkenazi Y-chromosome lineages of European descent.
Like Y23115 for example.

There where periods of time, when Jewish communities in Christian Europe where rich and well established, it could happen that there where male Christians who wanted to convert.
But we now understand that both the convert and Jews, who accepted him, would prefer to keep it secret. So, no doubt there will be no historical records about such facts. We know about a few cases of Christians executed tor an attempt to convert to Judaism, but this is the case when we only know about those who failed to keep it secret.
That is true that more converts were female, but we should not overestimate the percentage. If let's say 20-25% of converts where male this also gives us a significant number.

I'm sure it's not the only case, I know there are also very minor occurring West European (as in - German) paternal subclades that also exist in a very low percentages among Ashkenazi Jews.

However, there is no recorded history of something like this being common, as opposed to, for example, the detailed records of Eastern European women converting willingly to Judaism during the Reformation period and even slightly later. As a matter of fact, due to such cases, right after the Reformation, there were decrees by Rabbis harshly forbidding such conversions - this is something we haven't seen in preceding centuries.

Also, seeing how that R1a-YP1013 TMRCA is 750 CE, we can't also rule out Slavic slaves, which were beginning to introduced during that time period (8th century CE) and were still non-Christian:


Jews were one of the few groups who could move and trade between the Christian and Islamic worlds.[34] Ibn Khordadbeh observed and recorded routes of Jewish merchants in his "Book of Roads and Kingdoms" from the South of France to Spain, carrying (amongst other things) female slaves, eunuch slaves, and young slave boys. He also notes Jews purchasing Slavic slaves in Prague.

^That book was written in the 9th century CE.

And also:


Letters of Agobard, archbishop of Lyons (816–840),[36][37][38][39] acts of the emperor Louis the Pious,[40][41] and the seventy-fifth canon of the Council of Meaux of 845 confirms the existence of a route used by Jewish traders with Slavic slaves through the Alps to Lyon, to Southern France, to Spain.

And:


As German rulers of Saxon dynasties took over the enslavement (and slave trade) of Slavs in the 10th century, Jewish merchants bought slaves at the Elbe, sending caravans into the valley of the Rhine

https://en.wikipedia.org/wiki/Slavery_in_medieval_Europe#Jewish_merchants

And a reminder about Jewish law regarding slavery:


Christian leaders, including Pope Gregory the Great (served 590-604), objected to Jewish ownership of Christian slaves, due to concerns about conversion to Judaism and the Talmudic mandate to circumcise slaves.[87] The first prohibition of Jews owning Christian slaves was made by Constantine I in the 4th century. The prohibition was repeated by subsequent councils: Fourth Council of Orléans (541), Paris (633), Fourth Council of Toledo (633), the Synod of Szabolcs (1092) extended the prohibition to Hungary, Ghent (1112), Narbonne (1227), Béziers (1246). It was part of Saint Benedict's rule that Christian slaves were not to serve Jews.

https://en.wikipedia.org/wiki/Jewish_views_on_slavery#Post-Talmud_to_1800s

So while most slaves for Jews would also be mostly women, as a result of the fact that Jews in Medieval Europe didn't own lands and so would mostly use house slaves so, again, either women or eunuchs, one cannot also dismiss the possibility that some male Slavic slaves were also used, thus converted and eventually becoming part of the Jewish community.

This surely would fit a pre-13th century TMRCA (when Jews began settling en-mass in Eastern Europe).

hartaisarlag
07-30-2019, 09:11 PM
I believe a Jewish friend of mine (well, Jewish paternal grandfather) has that haplogroup (AFAIK, it's the only attested Ashkenazi branch of I-M423)—it was a bit strange to see that result when he first got his 23andMe back. Originally Balkan, it would seem? Heavy Greek skew; TMRCA of all Greek branches before the entry of Slavs into Greece.

Meanwhile, the second-largest Ashkenazi R1b lineage, R-A11720, looks solidly Balkan too. No Greeks, but plenty of Albanians.

Erikl86
07-30-2019, 09:24 PM
Meanwhile, the second-largest Ashkenazi R1b lineage, R-A11720, looks solidly Balkan too. No Greeks, but plenty of Albanians.

If indeed some of the non-Levantine paternal lineages in relatively significant numbers (still a minority though, as there is obvious prominence of Levantine/Near Eastern origin) will turn out to be Balkan with TMRCA of 300 BCE to 200 CE, it can also be a good proof for a more substantial East Mediterranean admixture than North Italian one, for instance.

artemv
07-30-2019, 09:35 PM
I'm sure it's not the only case, I know there are also very minor occurring West European (as in - German) paternal subclades that also exist in a very low percentages among Ashkenazi Jews.

However, there is no recorded history of something like this being common, as opposed to, for example, the detailed records of Eastern European women converting willingly to Judaism during the Reformation period and even slightly later. As a matter of fact, due to such cases, right after the Reformation, there were decrees by Rabbis harshly forbidding such conversions - this is something we haven't seen in preceding centuries.

Subbotnics, for example, were about 50% male.
At this period, conversions of females was somewhere more important than conversion of males - because of matrilinear definition of a Jew. So, female conversions might seem to be something more important for Rabbis.
But here we talk about something that happened much earlier than Subbotnics or Reformation period.


Also, seeing how that R1a-YP1013 TMRCA is 750 CE, we can't also rule out Slavic slaves, which were beginning to introduced during that time period (8th century CE) and were still non-Christian:

About R1a-YP1013, I see on yfull it has TMRCA of 1900 ybp. One of its subbranches, YP4848 has TMRCA 750 CE couldn't you mix this? Is all the YP1013 Jewish?

Thank you for this idea (about conversion to Judaism after being Jewish slaves), didn't think about this possibility.

But if we return to Y23115, its time estimations also seem to be too early.

I always agreed that Y3120 (TMRCA 2200 ybp, one of higher level branches of Y23115) is Slavic.
It has 4 main branches, and everything is quite fine with three of them - they are quite common in both Balkan and East/West Slav countries.
The fourth is Y18331 (TMRCA 2100 ybp).
This is a mostly Greek branch with a too early TMRCA.
I always thought that Y18331 was present in Greece because of 6th century Slav invasion to Greece. But now yfull recalculated the TMRCA and I started to doubt.

The problem with your idea (Y23115 captured from Slavs into slavery, sold to Jews, later got freed and joined Jewish community) is that diasporal Jewish Y23115 was formed too early (2100 ybp), maybe in some proto-Slav community, and that all its sibling clades (also formed 2100 ypb) ended up Greek, except for a single one Chuvash (non-Slavic East European). Now Chuvashia is part of Russia, so you will wee Russian flag (on y-full for example), but do not be mistaken, even about this case we cannot be sure about Slavic heritage.
In Greece y-full shows locations both on Peloponess and in Greek Macedonia (so, all on the continent).

Given the formation date, Y23115 could be present in Greece before it became Christian (or even before the Temple was destroyed) and convert to Judaism when this was still mainstream.
What do you think?

Agamemnon
07-30-2019, 10:09 PM
IMO Y18331 is definitely Slavic, as is Y3120 as a whole. The TMRCA estimates for the Greek branches are not incompatible with an arrival of the Slavs in Greece for that matter. The fact that A10959 is essentially Greek and Jewish IMO does give more weight to recent Aegean ancestry as I find it doubtful that this branch was picked up in Eastern Europe, it's certainly interesting to ponder the possibility that not all of the Slavic admixture in Ashkenazim came from Eastern Europe.

Y3120 is a fascinating lineage to be honest, it achieved tremendous reproductive success with the Slavic dispersals and was undoubtedly present among the Proto-Slavs. Though I am only superficially familiar with the question of Slavic ethnogenesis, I think the most logical sequence in archeological terms would be Przeworsk-Zarubintsy > Kiev > Prague-Korchak-Penkovka-Kolochin, Proto-Slavic is bound to have been spoken across material/cultural lines (so we should not seek a strict 1:1 correlation between language and material culture). I would place the Slavic homeland in or around the Pripet marshes. The presence of CTS10228* in Alsace makes a Western provenance quite likely, given the TMRCA estimates the move eastward might well be tied to Urnfield (yet again it's unlikely we'll be able to tell since Urnfield practiced cremation), it's one of those extreme cases where a single lineage that was relatively uncommon yet deeply-rooted in Europe experienced a demographic explosion (I1-M253 is a similar case).

hartaisarlag
07-30-2019, 10:21 PM
IMO Y18331 is definitely Slavic, as is Y3120 as a whole. The TMRCA estimates for the Greek branches are not incompatible with an arrival of the Slavs in Greece for that matter. The fact that A10959 is essentially Greek and Jewish IMO does give more weight to recent Aegean ancestry as I find it doubtful that this branch was picked up in Eastern Europe, it's certainly interesting to ponder the possibility that not all of the Slavic admixture in Ashkenazim came from Eastern Europe.

Y3120 is a fascinating lineage to be honest, it achieved tremendous reproductive success with the Slavic dispersals and was undoubtedly present among the Proto-Slavs. Though I am only superficially familiar with the question of Slavic ethnogenesis, I think the most logical sequence in archeological terms would be Przeworsk-Zarubintsy > Kiev > Prague-Korchak-Penkovka-Kolochin, Proto-Slavic is bound to have been spoken across material/cultural lines (so we should not seek a strict 1:1 correlation between language and material culture). I would place the Slavic homeland in or around the Pripet marshes. The presence of CTS10228* in Alsace makes a Western provenance quite likely, given the TMRCA estimates the move eastward might well be tied to Urnfield (yet again it's unlikely we'll be able to tell since Urnfield practiced cremation), it's one of those extreme cases where a single lineage that was relatively uncommon yet deeply-rooted in Europe experienced a demographic explosion (I1-M253 is a similar case).

Good observations. Going even further up the tree from CTS10228 reinforces the old Western/Central European provenance, and makes it likely that its path is Old W/C European > Slavic > Greek > Ashkenazi. Damn, strange. But upon examination, lots of lines are this strange.

Agamemnon
07-30-2019, 10:26 PM
Good observations. Going even farther up the tree from CTS10228 reinforces the old Western/Central European provenance, and makes it likely that its path is Old W/C European > Slavic > Greek > Ashkenazi. Damn, strange. But upon examination, lots of lines are this strange.

What strikes me is that this level of complexity under I-M170 is similar to what we find under J-M304. Even funnier is that there are several parallels between J1 and I1 on the one hand and J2 and I2 on the other.

hartaisarlag
07-31-2019, 02:55 AM
What strikes me is that this level of complexity under I-M170 is similar to what we find under J-M304. Even funnier is that there are several parallels between J1 and I1 on the one hand and J2 and I2 on the other.

In what sense?

artemv
07-31-2019, 04:20 AM
IMO Y18331 is definitely Slavic, as is Y3120 as a whole. The TMRCA estimates for the Greek branches are not incompatible with an arrival of the Slavs in Greece for that matter. The fact that A10959 is essentially Greek and Jewish IMO does give more weight to recent Aegean ancestry as I find it doubtful that this branch was picked up in Eastern Europe, it's certainly interesting to ponder the possibility that not all of the Slavic admixture in Ashkenazim came from Eastern Europe.

Y3120 is a fascinating lineage to be honest, it achieved tremendous reproductive success with the Slavic dispersals and was undoubtedly present among the Proto-Slavs. Though I am only superficially familiar with the question of Slavic ethnogenesis, I think the most logical sequence in archeological terms would be Przeworsk-Zarubintsy > Kiev > Prague-Korchak-Penkovka-Kolochin, Proto-Slavic is bound to have been spoken across material/cultural lines (so we should not seek a strict 1:1 correlation between language and material culture). I would place the Slavic homeland in or around the Pripet marshes. The presence of CTS10228* in Alsace makes a Western provenance quite likely, given the TMRCA estimates the move eastward might well be tied to Urnfield (yet again it's unlikely we'll be able to tell since Urnfield practiced cremation), it's one of those extreme cases where a single lineage that was relatively uncommon yet deeply-rooted in Europe experienced a demographic explosion (I1-M253 is a similar case).

Thank you, Aga.
But I meant something a bit different.
When I wrote Slavic, I meant the group of Early Slavs, just before their expansion, that started about the beginning of 6th century CE (maybe at the very end of 5th century CE).
So, my previous version was that Y18331 were a group of Slavs that invaded Greece in the end of 6th century, but given the early TMRCA of Y18331 I stared to doubt it. One more arguement against this scenario is that all the other three main branches of Y3120 are not concentrated at some Slavic countries, but are distributed among all the Slavic countries and have very few branches in Greece.

So I thought it could happen, that Y18331 left some pre-Slavic community before Early Slav expansions (and in this sence be "not Slavic"). And somehow got to Greece as a group of people - relatives by male line. This should have happen somewhere between 100 BCE-400 CE. Not sure they could get here as a group of slaves (although I cannot fully rule out this possibility). Just because a group of slaves would be probably sold to different parts of the Empire and because its difficult for slave Y-chromosome happlogroup to expand. Maybe as a group of legionaires of barbarian descent or confederates settled in Greece. The question is first of all to those who know history of Greece within Roman empire. Because proto-Slav history spesialist would not be able to tell us much - they have no aDNA, very few written sources (almost all of them are Greek/Roman), pure archeology. It's a bit boring to read their works with comparison and classification of hundreds of pottery vessels. No comparisons, we have much-much more information about Greece during Roman Empire period.

Yes, generally speaking, I very much doubt that Jews could get Y23115 in Ashkenazi time East Europe.
The same question is important about R1a-YP1013/YP4848. If all the R1a-YP1013 is Jewish, not just YP4848, given its early TMRCA it could be only picked up in the Mediterrian. In this case there is no sence to call R1a-YP1013 "Slavic". It was present in some pre-Slavic community, but left it before the Slavic times. If only YP4848 is Jewish, it changes the situation a little bit, in this case it could be Slavic. But I guess, even in this case it was likely picked up by the Jews in the Mediterrian also.

P.S. Wait a minute. Lets think about European mt-DNA in Ashkenazi Jewish gene pool in general. What if significant part of Ashkenazi Jewish mt-happlogroups comes from Roman Empire era slaves? If Jews picked up mt-DNA mostly from Greece or mostly from Italy, we would have already understood this. But Roman Empire era slaves were from all parts of Europe (and not just Europe), they were more often female, than male, finally, they must have been treated as Jews if they are freed. What do you think?

josh w.
07-31-2019, 01:18 PM
What's the subclade?

Go to the J2a Project. Within the Colored charts it is on pages 10 and 11. The SNPs in purple are PF 4610.....S23560 and Z43501? (The Caucasian 'cousins' are in orange on page 10.)

josh w.
07-31-2019, 04:11 PM
Go to the J2a Project. Within the Colored charts it is on pages 10 and 11. The SNPs in purple are PF 4610.....S23560 and Z43501? (The Caucasian 'cousins' are in orange on page 10.)

I doubt if my J2a Project sample is representative given the over-representation of Jews in western samples. This makes it difficult to estimate a time to the common ancestor.

josh w.
08-01-2019, 12:31 PM
I doubt if my J2a Project sample is representative given the over-representation of Jews in western samples. This makes it difficult to estimate a time to the common ancestor.

The J2a Project has posted a recent tree involving my subclade. It is hypothetical in the sense that some key ancestral steps are null categories-- they no longer have living members. My ancestral line was separated from the Caucasian lines by the mutation Z5048. No information on where or when the mutation appeared. About four 'generations' later my subclade appeared. There are no longer any members of the steps between Z 5048 and my subclade. There is no indication of when and where the mostly Jewish subclade first appeared. Again, a key problem is that there are few Ftdna members from the route between the mouth of the Euphrates and the Levant.

Under the old nomenclature I am J2a1b1a1a2 ( a branch of J2a*)

hartaisarlag
08-03-2019, 10:17 PM
Addendum (could go in this thread or another, but this is the busier one): the one confirmed E-Y6926* individual, upstream of Y6923 with a likely 2nd millennium BCE TMRCA between the 2 branches, can now be identified by the surname Mohammed. No country of origin specified yet, but to reiterate, he is a very close STR match with an individual named Al-Salman, and a cluster of related Omanis and Emiratis from the Al-Bulushi clan (of fabled Azd origin).

StillWater
08-05-2019, 02:26 AM
Excellent thread ! :)



Which one is the Levite one? IMO, this entered to Western Jews following the Geonic period of scholars from Babylonia settling in Spain and Ashkenaz, teaching Talmud.

This is what you may be looking for: https://avotaynuonline.com/2016/07/identifying-the-genetic-fingerprint-of-a-tzaddik-that-touched-the-world-the-shpoler-zeida/


The finding that the modal allele values reported at 37 STR markers for the pedigreed Zeida descendants in this study match those reported by for a small cluster of Ashkenazi non-Levite Jews that is not closely related to the larger Levite R1a group, and the associated hypothesis that such small clusters are often an indication that the most recent common ancestor arrived late in the Ashkenazi countries directly from the Middle East, is an interesting premise that bears further study.


Their haplotype, in conjunction with their haplogroup classification, R1a-M173>Z93>Z94>Y2632, represents the Y-DNA genetic signature of the Shpoler Zeida.

hartaisarlag
08-05-2019, 03:21 AM
This is what you may be looking for: https://avotaynuonline.com/2016/07/identifying-the-genetic-fingerprint-of-a-tzaddik-that-touched-the-world-the-shpoler-zeida/

Good find. A similar picture to R2a-FGC13211 (https://jewishdna.net/AB-085.html), which is not Levite. If anyone knows of other Ashkenazi Y-lines that appear to be of Mizrahi origins (whether Irano-Mesopotamian or Levantine in ultimate origin), I'd be curious. So far, that R2a branch is the only one of the 25 largest Ashkenazi branches with an apparent Mizrahi origin.

StillWater
08-05-2019, 03:40 AM
Good find. A similar picture to R2a-FGC13211 (https://jewishdna.net/AB-085.html), which is not Levite. If anyone knows of other Ashkenazi Y-lines that appear to be of Mizrahi origins (whether Irano-Mesopotamian or Levantine in ultimate origin), I'd be curious. So far, that R2a branch is the only one of the 25 largest Ashkenazi branches with an apparent Mizrahi origin.

The site has numerous interesting DNA articles. It seems that the Chassidic Twersky dynasty is almost surely Sephardic.

hartaisarlag
08-05-2019, 03:43 AM
The site has numerous interesting DNA articles. It seems that the Chassidic Twersky dynasty is almost surely Sephardic.

I’ll have to check that out! I would take Avotaynu’s interpretations with a grain of salt, same for JewishDNA.net, which is overdue for an update. To tests hypotheses like these, it’s important to search the projects and the latest trees.

Erikl86
08-05-2019, 04:53 AM
This is what you may be looking for: https://avotaynuonline.com/2016/07/identifying-the-genetic-fingerprint-of-a-tzaddik-that-touched-the-world-the-shpoler-zeida/

Thanks. Yes this is precisely what I was looking for.

I know it's a uniparental thread, but one thing which IMO should be noted though is that until recently, when we got both the aDNA for Roman-era Levantines AND IA Iranians, it was quite difficult to find any autosomal remnants for Mizrahi admixture in Western Jews - hell even now with the models showing ~10% IA Iranian admixture in Ashkenazim and Sephardim, I'm not so sure these are great models (best ones would be actual 5th century BCE Mesopotamian aDNA samples of course), but it's nice to see how the two - autosomal and uniparental - support each other - obviously I would expect that we'll see more paternal lineages coming from Mesopotamia (and almost no maternal, btw, because a Geonic-motivated movement of Mizrahi Jews to teach and spread Mishnah and Talmud in Western Jewish communities is bound to be predominantly male based), but it also gives a nicely historically plausible scenario for the over representation of such uniparentals among Ashkenazi Levites.

StillWater
08-05-2019, 05:06 AM
Thanks. Yes this is precisely what I was looking for.

I know it's a uniparental thread, but one thing which IMO should be noted though is that until recently, when we got both the aDNA for Roman-era Levantines AND IA Iranians, it was quite difficult to find any autosomal remnants for Mizrahi admixture in Western Jews - hell even now with the models showing ~10% IA Iranian admixture in Ashkenazim and Sephardim, I'm not so sure these are great models (best ones would be actual 5th century BCE Mesopotamian aDNA samples of course), but it's nice to see how the two - autosomal and uniparental - support each other - obviously I would expect that we'll see more paternal lineages coming from Mesopotamia (and almost no maternal, btw, because a Geonic-motivated movement of Mizrahi Jews to teach and spread Mishnah and Talmud in Western Jewish communities is bound to be predominantly male based), but it also gives a nicely historically plausible scenario for the over representation of such uniparentals among Ashkenazi Levites.

Do you mind linking the calculator you made for this? I'd like to run it on myself. Please tell me whether you used scaled/unscaled coordinates and what the penalty was.

coffeeprince
08-05-2019, 04:46 PM
Great thread, not much too add. I belong to a rare subclade below J2-L24, with a MRCA 2000 years. There's only a few people that have it besides me, a Greek, Italian, Puerto Rican, German and a Brit. Looks like it may have spread with the Roman empire.

hartaisarlag
08-05-2019, 04:55 PM
Great thread, not much too add. I belong to a rare subclade below J2-L24, with a MRCA 2000 years. There's only a few people that have it besides me, a Greek, Italian, Puerto Rican, German and a Brit. Looks like it may have spread with the Roman empire.

This is on your Ashkenazi side?

coffeeprince
08-05-2019, 05:07 PM
This is on your Ashkenazi side?

Sephardic side.

hartaisarlag
08-05-2019, 05:15 PM
Sephardic side.

Which is Moroccan, IIRC? Could you send me a link to the specific subclade on YFull?

I know there was recently a thread about Moroccan Jewish Y-haplogroups on The Apricity, which painted a pretty radical picture. It's a shame there isn't a good, transparent catalogue of Sephardic branches and their relative sizes (most branches are probably represented at JewishDNA.net, but scattered and poorly characterized).

coffeeprince
08-05-2019, 06:04 PM
Which is Moroccan, IIRC? Could you send me a link to the specific subclade on YFull?

I know there was recently a thread about Moroccan Jewish Y-haplogroups on The Apricity, which painted a pretty radical picture. It's a shame there isn't a good, transparent catalogue of Sephardic branches and their relative sizes (most branches are probably represented at JewishDNA.net, but scattered and poorly characterized).

Yes, correct.

Here it is: https://www.yfull.com/tree/J-PF2254/
My subclade is equivalent to the link I posted above.

hartaisarlag
09-09-2019, 12:09 AM
Bump. A summary worth adding, after some thought on the topic:

The main digest of Ashkenazi Y-DNA branches is informed by a bias toward origins in Spain (generally, in its early Jewish communities) for many Ashkenazi lines. I think this is ill-informed. It is indeed exciting, and in many cases possible, to identify a connection between an Ashkenazi line and Iberia, but conclusions about the nature of that connection should depend on the internal structure of the lineage.

5 or 6 out of 10 of the lines I mentioned above do not have an identifiable connection to Iberia, or to any non-Ashkenazi Jewish communities. With respect to the latter, I think it's totally possible that we'll find these connections with larger samples. These are R1a-Y2619, G2b-Y12975, Q1b-Y2200, E1b-Y14891, and J2a-Y15223; J2a-L556 is just barely in this category for the time being (though it has a branch, nested beneath an Ashkenazi line, that includes a Bulgarian from Ruse and an Italian from Veneto—not sure if these individuals are Jewish or not—plus a Honduran, a Mexican, and a Chilean not assigned to downstream branches).

4 of the 10 do have an identifiable to connection to either Iberia, non-Ashkenazi Jewish communities, or both. However, the stories of these connections appear to be very different for each of the 4: E1b-Y6923, J1-L816, J1-Z18271, and R1b-Y5051.

E1b-Y6923 seems to be the most straightforward: ca. 400 CE, it splits into two branches. One is exclusively Ashkenazi, and one is exclusively found in Hispanics and North African Jews (one of whom is possibly of Turkish Jewish origin; the post-1492 Sephardic origin of the other families is completely up in the air). My project has been trying to figure out what where the late classical common ancestor of these two sibling branches lived; my best guess is Italy, but there are other possibilities. (Here, the main variable of interest is the bidirectional flow of Jews between North Africa and Iberia between the 8th and 16th centuries CE.)

J1-L816 splits into two branches, ca. 850 CE. One is attested in a Spaniard, two Mexicans, and a German Jew; the other splits ca. 900 CE between a Greek Jewish branch (with a surname indicating Italian origin) and a large Ashkenazi branch. A Syrian Jew appears somewhere in the clade, but his location hasn't been confirmed, and a Spanish flag appears in the Greek Jewish branch, but I haven't found a record of its bearer. The divide here is much later, and more structured, than what we see in E-Y6923, which leaves open the possibility of a Spanish Jewish origin for the line, or a fairly late simultaneous French or Italian Jewish migration to both Ashkenaz and Sepharad.

J1-Z18271, as Agamemnon clarified earlier, is unique among Jewish branches because its early-middle 1st millennium BCE MRCA appears to have descendants all over the Jewish world, including Yemen and Iraq. Its main Ashkenazi branch seems to include some North African Jews too.

Finally, R1b-Y5051, despite appearing to have ultimately English/British roots, split ca. 500 CE between an Iberian branch (Spanish, Mexican, Portuguese, and Brazilian) and an Ashkenazi branch. Whether this indicates a path through Spain into the early Ashkenazi community or not is hard to say, but no non-Iberian continental European relatives of this line have been found sooner than 1000 BCE—which suggests that the original gentile ancestor of this line was actually Iberian.

If I'm wrong about any of this, please correct me. Also, please feel free to bounce ideas off me.

hartaisarlag
09-09-2019, 01:33 AM
Also: if anyone has any idea when Avotaynu plans on sharing its project results (especially non-Ashkenazi members') with the world, I'd love to know. They're sitting on top of quite a bit of data, keeping it completely inaccessible.

josh w.
09-09-2019, 11:30 PM
Also: if anyone has any idea when Avotaynu plans on sharing its project results (especially non-Ashkenazi members') with the world, I'd love to know. They're sitting on top of quite a bit of data, keeping it completely inaccessible.

I tried to contact them twice on my J2a* line but no response. I have evidence indicating that the line went from Tunisia to Iberia to the rest of Europe and Latin America. The director is familiar with that line personally.

hartaisarlag
09-10-2019, 12:09 AM
I tried to contact them twice on my J2a* line but no response. I have evidence indicating that the line went from Tunisia to Iberia to the rest of Europe and Latin America. The director is familiar with that line personally.

A friend through this stuff tells me they've hinted at an important role for North Africa in the Western Jewish ethnogenesis, but they've kept totally mum. I don't even know enough to know whether I generally trust their interpretations!

Claudio
09-10-2019, 01:10 AM
A friend through this stuff tells me they've hinted at an important role for North Africa in the Western Jewish ethnogenesis, but they've kept totally mum. I don't even know enough to know whether I generally trust their interpretations!

Any idea what there theory surmised?

hartaisarlag
09-10-2019, 01:19 AM
Any idea what there theory surmised?

Not a clue.

josh w.
09-10-2019, 05:31 PM
A friend through this stuff tells me they've hinted at an important role for North Africa in the Western Jewish ethnogenesis, but they've kept totally mum. I don't even know enough to know whether I generally trust their interpretations!

That was part of my problem. Avotaynu listed a Y dna match and surname with origins in Tunis. There was no response when I asked for more information. I also have suggestive information leading to the same conclusion regarding a different name or line. My dna ancestral families where quite famous (I don't know about my direct line). Wikipedia implied that the line was from Tunisia. It had an alternative name of Abulafia in Spain. I also have adna matches with the same family. They may be North African lines (Berber, Arabic) admixed into the Jewish population

Regarding Tunisia, I suspect that Jews gradually migrated west from Egypt and Israel as a result of the Diaspora. Punic migrations were less likely.

josh w.
09-10-2019, 07:12 PM
That was part of my problem. Avotaynu listed a Y dna match and surname with origins in Tunis. There was no response when I asked for more information. I also have suggestive information leading to the same conclusion regarding a different name or line. My dna ancestral families where quite famous (I don't know about my direct line). Wikipedia implied that the line was from Tunisia. It had an alternative name of Abulafia in Spain. I also have adna matches with the same family. They may be North African lines (Berber, Arabic) admixed into the Jewish population

Regarding Tunisia, I suspect that Jews gradually migrated west from Egypt and Israel as a result of the Diaspora. Punic migrations were less likely.

The Avotaynu listing indicated the line was in Tunis by 600 AD. I don't think that it came from Europe.

hartaisarlag
09-10-2019, 07:18 PM
The Avotaynu listing indicated the line was in Tunis by 600 AD. I don't think that it came from Europe.

Can you tell me which Avotaynu listing it is? Having trouble navigating to the right FTDNA project (it doesn't look like there's a general J2a one), and wading through these very slow-loading results pages.

josh w.
09-11-2019, 12:15 AM
Can you tell me which Avotaynu listing it is? Having trouble navigating to the right FTDNA project (it doesn't look like there's a general J2a one), and wading through these very slow-loading results pages.

Sorry, sorry. The Project is now M172. The SNP should be Z6048. The colored chart for M172 should now list my subclade pages 10-11

Except for Iberian surnames, there does not appear a non Jewish European Med cluster. This is why I do not think my Y line came to Tunisia from Europe. Tunisian Jewish lore does not point to Europe.

hartaisarlag
09-11-2019, 08:26 PM
Sorry, sorry. The Project is now M172. The SNP should be Z6048. The colored chart for M172 should now list my subclade pages 10-11

Except for Iberian surnames, there does not appear a non Jewish European Med cluster. This is why I do not think my Y line came to Tunisia from Europe. Tunisian Jewish lore does not point to Europe.

I don't see a mention of the Tunisian Jewish connection. I see Italians and an Iraqi in the Jewish cluster, and then just a bit out, but still under Z6048, Anglos, French, Germans, Italian, a Romanian, a Hadhrami Yemeni, and a Punjabi.

josh w.
09-11-2019, 11:44 PM
I don't see a mention of the Tunisian Jewish connection. I see Italians and an Iraqi in the Jewish cluster, and then just a bit out, but still under Z6048, Anglos, French, Germans, Italian, a Romanian, a Hadhrami Yemeni, and a Punjabi.

It is listed as one of my Ftdna Y matches. M172 gives subclade information but does not have the complete Ftdna list. I am not sure how to show my Ftdna list. M172 also implies non tested SNPs. M172 shows my results but not that of the Tunisian. Searching for the Tiapa family of Spain and Mexico might provide some information.

Yes, there are other Z6048 non Caucasian subclades that are not downstream of my subclade.

hartaisarlag
10-16-2019, 04:01 PM
On the topic of Tunisian—Ashkenazi connections, a Tunisian Jew I've been corresponding with forms a parallel subbranch (together with a Hispanic individual) to one of the 3 notable, but small, Ashkenazi branches of R1b-Z2103.

33940

In this case, the Tunisian Jewish/Hispanic branch and the Ashkenazi branch part ways ca. 900 CE, and the Tunisian Jew and Hispanic individual part ways ca. 1100 CE—the most parsimonious story for this would be a Sephardic origin for the former branch, and a divergence between it and the Ashkenazi branch somewhere like southern France (though Italy, Spain, and the Rhineland—the main documented cluster is of families named Spier or Shapiro—can't be ruled out).

This aligns with the fact that my correspondent has a likely Italian-origin name, and the Livornese Jews of Tunis were of Spanish/Portuguese origin—despite having a very pre-Sephardic-seeming genetic profile, and very few family lines explicitly connected with the Grana.

andyouare
10-16-2019, 07:27 PM
What do you guys think of the Y-lineage suggested here:
https://www.familytreedna.com/groups/cohenbrazil/about/background
http://jewishbubba.blogspot.com/2019/06/zadok-high-priest-of-judah-and.html

Principe
10-16-2019, 10:16 PM
I can give quite a lot of information on the Y15223 branch, as I have over 132 Y25 matches, this pertains to the Avotanyu list as I have a perfect Y12 and a 23/25 Y match with an Egyptian Jew but doesn’t appear on Y37 meaning he deviates from the Ashkenazi branch, and thus represents a break or older branch.

There is also an older Turkish Jewish L210 branch, no BigY has been done so whether or not the samples belong to Y15223 or Y15222 is unknown they might be an older Z482 branch or represent a Z489 branch, further testing should be encouraged, I tried emailing them back in 2017 with no luck.

There also dozens of Latin American in the Y15223 branch as well, one did a BigY and belongs to J-Y31769 hopefully he transfers his results to Yfull. We have 3 snp confirmed Y15223 families coming from Latin America, I know 1 family is Mexican. We also have others including 1 coming from Puerto Rico, 1 coming from Peru, 1 coming from Panama, with another 4 with no listed country of origin or I can’t see it based on not putting a family tree. These are all of my Y12, Y25, Y37 matches, in total 11 separate Latin American families containing Y15223.

On the Y15222 level, there is one man from Yemen who belongs to this branch based on Y strs he is close probably at the Z30386 level that everyone seems to forget when speaking of this branch.

hartaisarlag
10-16-2019, 10:22 PM
What do you guys think of the Y-lineage suggested here:
https://www.familytreedna.com/groups/cohenbrazil/about/background
http://jewishbubba.blogspot.com/2019/06/zadok-high-priest-of-judah-and.html

I don't know much, but what I'm sure is that these people have no particular reason to associate this line with the character Zadok.

andyouare
10-17-2019, 10:57 AM
I don't know much, but what I'm sure is that these people have no particular reason to associate this line with the character Zadok.

Yea I haven't found any sources for that particular claim, I'm wondering if anyone has any info on that subclade being Jewish.

hartaisarlag
10-22-2019, 12:48 AM
Unfortunately the latest YFull update has left a lot hanging, but I encourage people with special interest in any of these lineages to take a look.

Curious to hear Agamemnon's (or anyone else's) perspective on the changes to J-Y3088's structure and dating—the current update pushes the TMRCA yet another 200 years out of sync with Behar et al. 2017's, dropping it in the period of the proto-Israelites. Curiously, J-Y5400 now falls under a new branch called J-S17446, TMRCA: 3100 ybp, whose basal member is a Mexican. Not sure what to make of that.

Interestingly, this recurring 3100 ybp date is now given as the TMRCA of E-Y6926—in other words, the date of divergence between the Jews of E-Y6923 and their Gulf kin at E-Y6926*. The implications of this, if it's to be trusted, are quite different from those of a divergence ca. 3700 ybp.

hartaisarlag
12-03-2019, 06:57 PM
2 (well, 1 and 2) Ashkenazi R1b lines have recently caught my attention as I've delved into several Tunisian Jewish mysteries.

Parallel medieval origin stories; different ancient origins.

R-FGC14600 (TMRCA: 500 ybp) is downstream of R-L584, the classic West Asian R1b branch (I'm sure others can say more about its origins than me). According to JewishDNA.net, it's rather uncommon—in fact, only the 3rd largest Ashkenazi R-L584 branch. Its sibling branch, R-Y49084 (TMRCA: 900 ybp), has been confirmed in one Tunisian Jew and one Hispanic individual, and all but confirmed in another two people of Iberian descent (one Spanish, one Portuguese). These two branches share a common ancestor, R-Y11410, with a TMRCA of 1100 ybp. In turn, this branch shares an ancestor (R-Y16852) with Palestinians, Iraqis, Lebanese, an Iron Age NW Iran individual, Mexicans, Armenians, Russians ca. 4100 ybp. It, like other Ashkenazi R-L584 branches, is almost certainly of deep Levantine/West Asian origin. Notably, among Ashkenazim, this line is attested in old Rhineland communities, and is associated with the Shapiro family, suggesting a link to the early Ashkenazi community of Speyer.

NOTE: The Tunisian Jew in question has a last name that is attested in Livorno, and his fairly recent TMRCA with a Hispanic individual makes it pretty clear that his line is of Sephardic, rather than pre-Sephardic origin.

R-FGC20765 (TMRCA: 800 ybp) and R-FGC20816 (TMRCA: 850 ybp) are downstream of R-DF27, the classic Iberian R1b branch. Combined, these qualify as the largest Ashkenazi R-DF27 branch. Their third sibling is R-Y31658 (TMRCA: 900 ybp), and has been been found among a Livornese Jew, Turkish Jews, a Mexican, and a Spaniard, as well as an Italian basal to these lines, and all but confirmed in a Bulgarian Jew. These three branches share a common ancestor, R-V3476, with a TMRCA of 1150 ybp. In turn, this branch shares a common ancestor (R-FGC20758) with a Portuguese and a Brazilian ca. 2300 ybp, and is nested within deep Iberian phylogeny.

Why do I bring these up? For one, because of how well-documented and star-shaped R1b is, it's pretty easy to pinpoint the origin of most Ashkenazi R1b lines, and these two couldn't be more different in their origins. But more importantly, we have two (two and a half?) examples of lines shared by Ashkenazim and Sephardim, which parted ways ca. 850 to 900 CE. Besides fitting the conventional line on the date of the proto-Ashkenazi community's founding, this date range matches the TMRCAs of a majority of the large Ashkenazi lines described. It also matches the date of the Ashkenazi-Hispanic divergence within Ashkenazi "big group" J-L816. On the other hand, the Ashkenazi—Sephardi/North African Jewish/Hispanic divergence within the big group E-Y6923 dates to several hundred years earlier.

What does a split between Ashkenazi and Sephardic lines ca. 850-900 CE entail? Well, in the case of R-V3476, which is ultimately Iberian in origin, this seems like a clear case of early Sephardic introgression into the proto-Ashkenazi community. The fact that 2 of 3 sibling branches are Ashkenazi, while only one remained Sephardi, complicates things a bit, but in any case, the picture of an Iberian > Sephardic > Ashkenazi introgression is unavoidable. Now, was this the result of multiple Spain > Germany migrations, or a trickle of related families coming up from a southern French community that was in close touch with the proto-Sephardic community?

R-Y11410, on the other hand, shows no sign of a primary origin among Sephardim and later introgression into Ashkenazim. But compared to, say, E-Y6923, the line's location of origin is heavily constrained by its 900 CE TMRCA. Southern France? Where else would have been supplying migrations to both the Rhineland and Spain around that date?

hartaisarlag
12-03-2019, 10:01 PM
In addition: the 3rd largest Ashkenazi R1b branch overall (considering these numbers from JewishDNA.net with a grain of salt), AB-069 / R1b-V88-FGC20980-FGC21047, fits this same pattern.

It's the largest Ashkenazi R1b branch of likely Near Eastern origin, first of all—its nearest relations, within a TMRCA of 3400 ybp, are Saudi, Maltese, and Sudanese.

R-FGC20980 splits ca. 1200 ybp into R-FGC21047, its Ashkenazi branch, and R-Y34349, a branch including two Mexicans, a Peruvian, and a Sephardi Jew from Salonica (with its own TMRCA of 1200 ybp). This is a direct parallel to the case of R-Y11410—a Near Eastern (Israelite, perhaps) origin, and then an Ashkenazi—Sephardi split, probably somewhere in between Spain and Germany, between 800 and 900 CE.

Granted, this branch only accounts for about 1% of Ashkenazi Y-DNA lineages. However, the pattern of Ashkenazi—Sephardi divergences in the 800-900 CE time frame suggests that the two populations have at least one partial founding stock in common rather late in the game. I'd be curious to hear about other examples of this late divergence, or for that matter, any clear examples besides my own line's, of earlier divergences (e.g. late classical, very early medieval).

hartaisarlag
12-05-2019, 08:43 PM
With the end of a semester in sight (and PhD apps done), I've come back to this line of investigation in earnest.

I'm not going to write a biographical sketch of every single Ashkenazi Y-lineage, but I have gone with a fine-toothed comb through the top 35, according to JewishDNA.net (which I now realize is a very iffy guide to relative frequency). In a brilliant illustration of universal power laws, these comprise 22% of all known Ashkenazi lines, but account for 81% of Ashkenazi individuals. All kinds of good stuff. I encourage commentary and pushback from anyone who knows more about a given line than me, or doubts my interpretations for any other reason.

I guess a good place to begin would be enumerating the branches that we can say for sure are the result of European introgression. Beyond that, the picture is murky in nearly every case, with one exciting new exception. Of the top 35, I count 6—but this is most likely an undercounting:

— R1b-U106 > Y5051 (/FGC8564), the Ivanhoe cluster [Germanic/Iberian]
— R1b-PF7558 > A11720
— R1b-U152 > L408
— R1b-U152 > BY187600 [Italian/German/French]
— I2-M423 > Y23115 [Greek/Slavic]
— J2b-L283 > Y33795, the J2b Kohen cluster [Etruscan/Italian]

— [B]R1b-U106 > Y5051 (/FGC8564), the Ivanhoe cluster
The largest clearly European-introgressed Ashkenazi Y branch (~2.9%), R1b-Y5051 remains a puzzle despite investigators' best efforts. Its ultimate origin is clearly Germanic, and directly upstream of it within a TMRCA of 1500 ybp are English and Swedish individuals; within a TMRCA timescale of 1950 ybp, you get more Englishmen, plus a German from Baden (just shy of the Rhineland). However, right after this split from the English and Swedish branch, Y5051 splits into a Spanish/Portuguese branch and an Ashkenazi branch (R1b-FGC8564). Note, there are no known Sephardim in this branch, despite the putative Iberian connection.

Does this structure imply a Germanic > Iberian > Ashkenazi introgression? Despite theories about a British/English connection, the date perfectly fits a Visigothic arrival in Iberia. But is that at all connected to the lineage's presence in Ashkenazim? For now I think it's too early to say; if bona fide Sephardim turn up, we can assume so, but if closer German matches turn up, I'm inclined to chalk this up to a Rhenish convert. A problem for the latter theory is this branch's TMRCA of 500 CE (that is, it begins to rapidly split almost simultaneously with its split from the Iberian branch, and in turn with its split from the English-Swedish branch). This date isn't anomalous for an Ashkenazi founding line TMRCA (Q-Y2200, J2a-Y15223, and J1-Y5400, with its own idiosyncrasies, are in close range of it)—but it suggests that its incorporation into the Jewish population and its subsequent branching would have begun hundreds of years before the founding of the proto-Ashkenazi community in the Rhineland.

I'd love to hear more from R-U106 experts on this!

— R1b-PF7558 > A11720
R1b-A11720 (~1.7%) is also a difficult one to pin down, but by contrast, it's on a relatively understudied branch of R1b, and its nearest known sister branch shares with it a TMRCA of 4600 ybp. The branch itself has a TMRCA of 650 CE—relatively typical, but of little help in establishing the date of introgression. This sister branch, R1b-BY16680, seems to have a pan-Balkan distribution, with representation in Romania, Bulgaria, Serbia, and Kosovo/Albania; the FTDNA Haplotree also identifies two Italians and a Spaniard. Cousin branches, all falling under R-PF7563 (TMRCA: 4600 ybp) seem to be rooted in the Southern Balkans and Italy, with downstream presences in Lebanon and Saudi Arabia. This raises the possibility of some pre-exilic SE European introgression into the Ashkenazi gene pool, but without a closer Near Eastern link, I am inclined to chalk this up to introgression within Europe—either Italy or the Balkans. In any event, it is ultimately a European one.

— R1b-U152 > L408
Unfortunately, I can't give a TMRCA for this one, as YFull's U152 downstream structure is way too spare. A sister branch is found in an Italian line from Marche and a German line with deep roots in colonial America, whose surname is commonest in Baden-Württemberg and the Rhineland. One step further out are Italians from several regions, a French-Canadian, a Francophone Swiss man, as well as several English and English-origin Irish people. Someone who knows R-U152 > R-L2 better than I do (R. Rocca, cough) can comment, but it seems that despite an Italic origin, this broad grouping is pretty widely distributed within northwestern Europe. So an origin in Italy or Germany (or France) is possible; British seems historically untenable.

— R1b-U152 > BY187600
Similar issues as above. Parallel lines in Germany, England/Scotland(/Ireland), Luxembourg, Tatarstan, and Italy (Marche). No TMRCA calculation given. Origin in Italy or Germany (or France) is possible; British seems historically untenable.

— I2-M423 > Y23115
This one has been discussed before, and seems to be the best, most prominent candidate for Greek male introgression. Within a TMRCA of 2100 ybp, the branch has a Peloponnesian-Albanian sibling branch, and several steps up, it's mostly Greeks (with a Chuvash, a Russian, and a Ukrainian thrown in). The parent-branch's Greek history probably wasn't very deep before its entry into the pre-proto-Ashkenazi gene pool; its provenance is probably ultimately Old European > Balkan/Eastern European > Greek. AFAIK, the timeline is a bit early for a direct Slavic > Greek introgression, but northerly origins seem like a safe assumption. A caveat, though: there is a big gap between 100 BCE and the Ashkenazi branch's TMRCA of 1000 CE. It is theoretically possible that other non-Greek, non-Jewish Eastern European branches will show up under the immediate parent branch, I-A10959, leaving open the possibility of a Slavic introgression.

— J2b-L283 > Y33795, the J2b Kohen cluster
Despite a long history of wrangling over the origins of J2b-L283 as it appears in a small minority of Jews and Levantines, the recent Rome paper has gotten us close to a solution to this one. J2b-Y33975 falls directly under CTS6190 (TMRCA: 3100 ybp), which was found in an Iron Age Etruscan, and whose other descendants include Italians and Portuguese individuals. An ultimately Etruscan, and proximately Roman convert origin seems likely. The line's relatively early TMRCA of 200 CE is notable for two reasons: 1) it constrains the date of this conversion/introgression event to relatively early in Ashkenazi Jews' pre-Ashkenazi European history, and makes it likely that the line will show up in non-Ashkenazi Jews down the road, and 2) it resembles the early TMRCA's of other major priestly and Levite lines, suggesting that after the conversion event, this lineage's kohen status was preserved fervently, and most likely, across multiple communities.

EDIT: The Portuguese-flagged individual under sister-branch J2b-Y22038 (he shares it with a Tuscan, and an Italian with a surname mostly known from Liguria; TMRCA: 1900 ybp) is apparently a Sephardic Italian Jew of Portuguese origin, which implies, unless there is hidden structure under CTS6190 binding this group to the Ashkenazi group, that multiple CTS6190 individuals became Jewish in Roman-era Italy. I do not believe this individual identifies as a kohen.

One other line on another branch (J2b-Y36166), flagged as Portuguese, identifies as the descendant of an 11th-century Portuguese Jew, whose Caribbean Sephardic ancestor was adopted by an Anglo-Virginian in the early 19th century. There are too many questionable links in the story to fully credit it, but a Sephardic origin seems on the face of it likelier than an English one.

But wait, it gets stranger. The other Portuguese-flagged member of J2b-Y36166 (TMRCA: 225 ybp!) is [I]also an Anglo-Virginian; different surname, different part of the state, no known Sephardic, Italian, or Iberian roots. And under J-CTS6190*, yet a third Anglo-Virginian. What the hell?

StillWater
12-05-2019, 10:11 PM
With the end of a semester in sight (and PhD apps done), I've come back to this line of investigation in earnest.

I'm not going to write a biographical sketch of every single Ashkenazi Y-lineage, but I have gone with a fine-toothed comb through the top 35, according to JewishDNA.net (which I now realize is a very iffy guide to relative frequency). In a brilliant illustration of universal power laws, these comprise 22% of all known Ashkenazi lines, but account for 81% of Ashkenazi individuals. All kinds of good stuff. I encourage commentary and pushback from anyone who knows more about a given line than me, or doubts my interpretations for any other reason.

I guess a good place to begin would be enumerating the branches that we can say for sure are the result of European introgression. Beyond that, the picture is murky in nearly every case, with one exciting new exception. Of the top 35, I count 6—but this is most likely an undercounting:

— R1b-U106 > Y5051 (/FGC8564), the Ivanhoe cluster [Germanic/Iberian]
— R1b-PF7558 > A11720
— R1b-U152 > L408 [Italian/German/French]
— R1b-U152 > BY187600 [Italian/German/French]
— I2-M423 > Y23115 [Greek/Slavic]
— J2b-L283 > Y33795, the J2b Kohen cluster [Etruscan/Italian]

— [B]R1b-U106 > Y5051 (/FGC8564), the Ivanhoe cluster
The largest clearly European-introgressed Ashkenazi Y branch (~2.9%), R1b-Y5051 remains a puzzle despite investigators' best efforts. Its ultimate origin is clearly Germanic, and directly upstream of it within a TMRCA of 1500 ybp are English and Swedish individuals; within a TMRCA timescale of 1950 ybp, you get more Englishmen, plus a German from Baden (just shy of the Rhineland). However, right after this split from the English and Swedish branch, Y5051 splits into a Spanish/Portuguese branch and an Ashkenazi branch (R1b-FGC8564). Note, there are no known Sephardim in this branch, despite the putative Iberian connection.

Does this structure imply a Germanic > Iberian > Ashkenazi introgression? Despite theories about a British/English connection, the date perfectly fits a Visigothic arrival in Iberia. But is that at all connected to the lineage's presence in Ashkenazim? For now I think it's too early to say; if bona fide Sephardim turn up, we can assume so, but if closer German matches turn up, I'm inclined to chalk this up to a Rhenish convert. A problem for the latter theory is this branch's TMRCA of 500 CE (that is, it begins to rapidly split almost simultaneously with its split from the Iberian branch, and in turn with its split from the English-Swedish branch). This date isn't anomalous for an Ashkenazi founding line TMRCA (Q-Y2200, J2a-Y15223, and J1-Y5400, with its own idiosyncrasies, are in close range of it)—but it suggests that its incorporation into the Jewish population and its subsequent branching would have begun hundreds of years before the founding of the proto-Ashkenazi community in the Rhineland.

I'd love to hear more from R-U106 experts on this!

— R1b-PF7558 > A11720
R1b-A11720 (~1.7%) is also a difficult one to pin down, but by contrast, it's on a relatively understudied branch of R1b, and its nearest known sister branch shares with it a TMRCA of 4600 ybp. The branch itself has a TMRCA of 650 CE—relatively typical, but of little help in establishing the date of introgression. This sister branch, R1b-BY16680, seems to have a pan-Balkan distribution, with representation in Romania, Bulgaria, Serbia, and Kosovo/Albania; the FTDNA Haplotree also identifies two Italians and a Spaniard. Cousin branches, all falling under R-PF7563 (TMRCA: 4600 ybp) seem to be rooted in the Southern Balkans and Italy, with downstream presences in Lebanon and Saudi Arabia. This raises the possibility of some pre-exilic SE European introgression into the Ashkenazi gene pool, but without a closer Near Eastern link, I am inclined to chalk this up to introgression within Europe—either Italy or the Balkans. In any event, it is ultimately a European one.

— R1b-U152 > L408
Unfortunately, I can't give a TMRCA for this one, as YFull's U152 downstream structure is way too spare. A sister branch is found in an Italian line from Marche and a German line with deep roots in colonial America, whose surname is commonest in Baden-Württemberg and the Rhineland. One step further out are Italians from several regions, a French-Canadian, a Francophone Swiss man, as well as several English and English-origin Irish people. Someone who knows R-U152 > R-L2 better than I do (R. Rocca, cough) can comment, but it seems that despite an Italic origin, this broad grouping is pretty widely distributed within northwestern Europe. So an origin in Italy or Germany (or France) is possible; British seems historically untenable.

— R1b-U152 > BY187600
Similar issues as above. Parallel lines in Germany, England/Scotland(/Ireland), Luxembourg, Tatarstan, and Italy (Marche). No TMRCA calculation given. Origin in Italy or Germany (or France) is possible; British seems historically untenable.

— I2-M423 > Y23115
This one has been discussed before, and seems to be the best, most prominent candidate for Greek male introgression. Within a TMRCA of 2100 ybp, the branch has a Peloponnesian-Albanian sibling branch, and several steps up, it's mostly Greeks (with a Chuvash, a Russian, and a Ukrainian thrown in). The parent-branch's Greek history probably wasn't very deep before its entry into the pre-proto-Ashkenazi gene pool; its provenance is probably ultimately Old European > Balkan/Eastern European > Greek. AFAIK, the timeline is a bit early for a direct Slavic > Greek introgression, but northerly origins seem like a safe assumption. A caveat, though: there is a big gap between 100 BCE and the Ashkenazi branch's TMRCA of 1000 CE. It is theoretically possible that other non-Greek, non-Jewish Eastern European branches will show up under the immediate parent branch, I-A10959, leaving open the possibility of a Slavic introgression.

— J2b-L283 > Y33795, the J2b Kohen cluster
Despite a long history of wrangling over the origins of J2b-L283 as it appears in a small minority of Jews and Levantines, the recent Rome paper has gotten us close to a solution to this one. J2b-Y33975 falls directly under CTS6190 (TMRCA: 3100 ybp), which was found in an Iron Age Etruscan, and whose other descendants include Italians and Portuguese individuals. An ultimately Etruscan, and proximately Roman convert origin seems likely. The line's relatively early TMRCA of 200 CE is notable for two reasons: 1) it constrains the date of this conversion/introgression event to relatively early in Ashkenazi Jews' pre-Ashkenazi European history, and makes it likely that the line will show up in non-Ashkenazi Jews down the road, and 2) it resembles the early TMRCA's of other major priestly and Levite lines, suggesting that after the conversion event, this lineage's kohen status was preserved fervently, and most likely, across multiple communities.

EDIT: The Portuguese-flagged individual under sister-branch J2b-Y22038 (he shares it with a Tuscan...; TMRCA: 1900 ybp) is apparently a Sephardic Italian Jew of Portuguese origin, which implies, unless there is hidden structure under CTS6190 binding this group to the Ashkenazi group, that multiple CTS6190 individuals became Jewish in Roman-era Italy. I do not believe this individual identifies as a kohen.

I'm not entirely sold on J2b-Y33795 coming post-exile. I certainly see the argument for it. However, looking at J-PH1602, for example. What do we call it - CTS6190's sister branch? There is now a Syrian result among the European ones. This seems to be incredibly complex lineage, phylogenically. It seems to have hits all over the Mediterannian - north and south. I think we underestimate ancient travel around the Sea (i.e Sea Peoples). The main inspiration for my skepticism is the Jewish lineage E-MZ236. It seemed post-exilic to many, but then a basal Lebanese popped up right next to the Jewish branch, all with a "recent" TMRCA. Likewise, the lineages above, at least by YFull, seem to also be heavily Portuguese, as with J2b-Y33795. Coupled with the fact that we must now assume that multiple CTS6190 individuals became Jewish in Roman Italy, the likelihood for J2b-Y33795 being post-exilic, while considerable, isn't overwhelming. It may be just like E-MZ236 - a late arrival to Judea, and very possibly from Mediterranean Europe. Also, a Cohen lineage would've been much more crucial for social status in Judea than in Roman Italy.

John Doe
12-05-2019, 10:16 PM
What can you tell about E-Z830 and E-PF1975 in particular? From what I understand it's quite rare, which figures since I found almost no info about it anywhere

hartaisarlag
12-05-2019, 10:22 PM
I'm not entirely sold on J2b-Y33795 coming post-exile. I certainly see the argument for it. However, looking at J-PH1602, for example. What do we call it - CTS6190's sister branch? There is now a Syrian result among the European ones. This seems to be incredibly complex lineage, phylogenically. It seems to have hits all over the Mediterannian - north and south. I think we underestimate ancient travel around the Sea (i.e Sea Peoples). The main inspiration for my skepticism is the Jewish lineage E-MZ236. It seemed post-exilic to many, but then a basal Lebanese popped up right next to the Jewish branch, all with a "recent" TMRCA. Likewise, the lineages above, at least by YFull, seem to also be heavily Portuguese, as with J2b-Y33795. Coupled with the fact that we must now assume that multiple CTS6190 individuals became Jewish in Roman Italy, the likelihood for J2b-Y33795 being post-exilic, while considerable, isn't overwhelming. It may be just like E-MZ236 - a late arrival to Judea, and very possibly from Mediterranean Europe. Also, a Cohen lineage would've been much more crucial for social status in Judea than in Roman Italy.

Color me skeptical, re: the relevance of the new Syrian individual. YFull doesn't show it, but on the FTDNA haplotree, this Syrian is nested relatively deep in the J-PH1602 phylogeny, clustering with a Bosnian, a Brit, and a German, on a branch parallel to one with an American (likely British) and a German; these in turn are downstream of a few Germans, and all parallel branches are 100% European so far. I would assume the introgression into Syria is post-classical, based on this phylogeny. It would be more accurate to characterize J2b-Z2507 as having hits all over Europe—north and south, rather than all over the Mediterranean basin. That Syrian is the only exception.

hartaisarlag
12-05-2019, 10:30 PM
I'm not entirely sold on J2b-Y33795 coming post-exile. I certainly see the argument for it. However, looking at J-PH1602, for example. What do we call it - CTS6190's sister branch? There is now a Syrian result among the European ones. This seems to be incredibly complex lineage, phylogenically. It seems to have hits all over the Mediterannian - north and south. I think we underestimate ancient travel around the Sea (i.e Sea Peoples). The main inspiration for my skepticism is the Jewish lineage E-MZ236. It seemed post-exilic to many, but then a basal Lebanese popped up right next to the Jewish branch, all with a "recent" TMRCA. Likewise, the lineages above, at least by YFull, seem to also be heavily Portuguese, as with J2b-Y33795. Coupled with the fact that we must now assume that multiple CTS6190 individuals became Jewish in Roman Italy, the likelihood for J2b-Y33795 being post-exilic, while considerable, isn't overwhelming. It may be just like E-MZ236 - a late arrival to Judea, and very possibly from Mediterranean Europe. Also, a Cohen lineage would've been much more crucial for social status in Judea than in Roman Italy.

And again: the Italian Sephardic line shares a TMRCA with a Tuscan at ca. 100 CE; another Italian falls in this branch, and I can’t say how this affects the TMRCA. You’re right that the existence of multiple Jewish CTS6190 branches complicates things somewhat, but what I don’t think this suggests is a much earlier introgression, and then a bleeding out of originally Italian patrilines, absorbed by Jews in the Levant, back out into the Italian population.

hartaisarlag
12-05-2019, 11:03 PM
What can you tell about E-Z830 and E-PF1975 in particular? From what I understand it's quite rare, which figures since I found almost no info about it anywhere

I wish I could say more about the two E-Z830+ E-M123- Ashkenazi branches — they’re relatively large and one of my best friends is a member — but for now, they remain enigmatic.

Agamemnon
12-05-2019, 11:38 PM
I wish I could say more about the two E-Z830+ E-M123- Ashkenazi branches — they’re relatively large and one of my best friends is a member — but for now, they remain enigmatic.

Understatement. We're not only dealing with rare and discretely basal branches, they're looking more and more like deeply-rooted "survivor" lineages from the Epipaleolithic period similar to other lineages we know of in Europe that have comparable gaps between their formation and TMRCAs, lineages that are always (without fault) derived from Europe's Paleolithic population. In both cases, it's extremely tempting to assume that these lineages experienced their demographic peak during the Paleolithic (probably because of a combination of advantages that gave these groups an edge, such as better diet, innovative hunting techniques and new technology) and then went through a very long period of decline, explaining the large gaps I spoke of.

The entire dilemma here is that beyond an association with the Natufians, and probably the Kebarans before them, it's hard to pinpoint any informative patterns that could give us an idea of what their descendants were up to. Beyond basically having a very ancient presence in the Levant and having stayed put there, not much is certain.

hartaisarlag
12-06-2019, 01:27 AM
Understatement. We're not only dealing with rare and discretely basal branches, they're looking more and more like deeply-rooted "survivor" lineages from the Epipaleolithic period similar to other lineages we know of in Europe that have comparable gaps between their formation and TMRCAs, lineages that are always (without fault) derived from Europe's Paleolithic population. In both cases, it's extremely tempting to assume that these lineages experienced their demographic peak during the Paleolithic (probably because of a combination of advantages that gave these groups an edge, such as better diet, innovative hunting techniques and new technology) and then went through a very long period of decline, explaining the large gaps I spoke of.

The entire dilemma here is that beyond an association with the Natufians, and probably the Kebarans before them, it's hard to pinpoint any informative patterns that could give us an idea of what their descendants were up to. Beyond basically having a very ancient presence in the Levant and having stayed put there, not much is certain.

All that can be said is that the two branches seem to share a common ancestor in E-PF2025/E-PF1975 with a TMRCA of 9200 ybp, and that for one of them (E-PF1974), the Ashkenazi branch E-FGC56002 seems to be lodged downstream of some Sardinians, some Iraqis, and an Englishman. There are also a Brit and a Swede in the Ashkenazi branch, but that's not unheard of. Other than that, no other non-Ashkenazi presence attested in this entire sibling branch of E-M123! (Actually, checking now, there seems to be an Armenian from Diyarbakir and an Italian who are E-PF2025/E-PF1975 with no downstream hits on the Haplotree.)

Also, AbdoNumen claimed on another thread that Italian Jews have been found in this branch (if you could call such a deeply-diverged lineage a "branch") - can I get receipts? For that matter, any Italian Jewish uniparentals would be deeply appreciated.

hartaisarlag
12-06-2019, 04:33 PM
One other line on another branch (J2b-Y36166), flagged as Portuguese, identifies as the descendant of an 11th-century Portuguese Jew, whose Caribbean Sephardic ancestor was adopted by an Anglo-Virginian in the early 19th century. There are too many questionable links in the story to fully credit it, but a Sephardic origin seems on the face of it likelier than an English one.

But wait, it gets stranger. The other Portuguese-flagged member of J2b-Y36166 (TMRCA: 225 ybp!) is also an Anglo-Virginian; different surname, different part of the state, no known Sephardic, Italian, or Iberian roots. And under J-CTS6190*, yet a third Anglo-Virginian. What the hell?[/B]

Update: J2b-Y36166 has no actual Portuguese or Jewish connection, as far as I can tell. It is probably an English line that introgressed into Britain during Roman times.

One of the branch members [Harlow] appears to have fabricated a Caribbean Sephardic ancestor, and spun great tales of him online—but fortunately, it is harder and harder to successfully bullshit these days. Everyone's investigating. The other [Bowles] doesn't even appear to claim Portuguese or Sephardic ancestry—his paternal line goes back to 13th century Lincolnshire—so I'm not sure how he ended up with a Portuguese flag on YFull.

My browsing of Jewish-adjacent R-U152 lines shows me that this English presence on multiple branches is nothing unusual, as unintuitive as it might seem. But I think someone thought it'd be easy to hijack people's intuition that J-CTS6190 belongs with Sephardim, and not with English gentiles, and built a story around it.

AbdoNumen
12-07-2019, 02:06 AM
Also, AbdoNumen claimed on another thread that Italian Jews have been found in this branch (if you could call such a deeply-diverged lineage a "branch") - can I get receipts? For that matter, any Italian Jewish uniparentals would be deeply appreciated.

Behar et al 2017 (R1a among Ashkenazi Levites) has two Italian-Jewish lineages: One under E-Z830 -> PF1962 -> Y56257 (https://www.yfull.com/tree/E-Y56257/) (M123-) and the other under R-L23 -> L51 -> Y125239 (https://www.yfull.com/tree/R-Y125239/)

35204

hartaisarlag
12-07-2019, 03:35 AM
Behar et al 2017 (R1a among Ashkenazi Levites) has two Italian-Jewish lineages: One under E-Z830 -> PF1962 -> Y56257 (https://www.yfull.com/tree/E-Y56257/) (M123-) and the other under R-L23 -> L51 -> Y125239 (https://www.yfull.com/tree/R-Y125239/)

35204

Thank you. I think I'm sitting on a very gracious e-mail about Mountain Jewish lineages that you sent a few months ago, btw.

StillWater
12-07-2019, 05:08 AM
And again: the Italian Sephardic line shares a TMRCA with a Tuscan at ca. 100 CE; another Italian falls in this branch, and I can’t say how this affects the TMRCA. You’re right that the existence of multiple Jewish CTS6190 branches complicates things somewhat, but what I don’t think this suggests is a much earlier introgression, and then a bleeding out of originally Italian patrilines, absorbed by Jews in the Levant, back out into the Italian population.

I was over zealous in my argument for it. I certainly agree that I presented the less likely scenario. However, I don't think its chance is negligible based on its, at least superficial resemblance to E-MZ236. And while, a route from the North Mediterranean to Israel, and to Europe back again seems to break Occam's Razor, it does seem to be the case with E-MZ236; and given that it would've occurred over centuries, it's not so crazy. Also, this isn't the only possible route. It could be that the Etruscan got it from a more southern source. We did see, for example, Latin tribe outliers that were deemed to be half Phoenician.

hartaisarlag
12-07-2019, 06:37 AM
I was over zealous in my argument for it. I certainly agree that I presented the less likely scenario. However, I don't think its chance is negligible based on its, at least superficial resemblance to E-MZ236. And while, a route from the North Mediterranean to Israel, and to Europe back again seems to break Occam's Razor, it does seem to be the case with E-MZ236; and given that it would've occurred over centuries, it's not so crazy. Also, this isn't the only possible route. It could be that the Etruscan got it from a more southern source. We did see, for example, Latin tribe outliers that were deemed to be half Phoenician.

Very fair, and thanks for alerting me to E-MZ236’s history. All I’ll say is that E-M81 is structurally perplexing.

John Doe
12-09-2019, 11:08 PM
Just looked at this site http://scaledinnovation.com/gg/snpTracker.html?snp=E-FT111977,R-BY16172
Searched for E-PF1975 on it, it seems he decided to place this subclade's source around Greece. Kinda wondering on what data he was basing it

hartaisarlag
12-10-2019, 08:18 PM
Just looked at this site http://scaledinnovation.com/gg/snpTracker.html?snp=E-FT111977,R-BY16172
Searched for E-PF1975 on it, it seems he decided to place this subclade's source around Greece. Kinda wondering on what data he was basing it

Any averaged ancestor location data relying heavily on Jewish Diaspora samples will be necessarily skewed. I don’t think it’s anything more sophisticated than that.

artemv
12-12-2019, 12:30 PM
Just looked at this site http://scaledinnovation.com/gg/snpTracker.html?snp=E-FT111977,R-BY16172
Searched for E-PF1975 on it, it seems he decided to place this subclade's source around Greece. Kinda wondering on what data he was basing it

They do not have reliable info about many subclades. Tried to check what will it show for V88 - the results looks like they in scaledinnovation never new this is a European HG happlogroup, show some location in Negev desert.

hartaisarlag
12-16-2019, 08:08 PM
With the end of a semester in sight (and PhD apps done), I've come back to this line of investigation in earnest.

I'm not going to write a biographical sketch of every single Ashkenazi Y-lineage, but I have gone with a fine-toothed comb through the top 35, according to JewishDNA.net (which I now realize is a very iffy guide to relative frequency). In a brilliant illustration of universal power laws, these comprise 22% of all known Ashkenazi lines, but account for 81% of Ashkenazi individuals. All kinds of good stuff. I encourage commentary and pushback from anyone who knows more about a given line than me, or doubts my interpretations for any other reason.

I guess a good place to begin would be enumerating the branches that we can say for sure are the result of European introgression. Beyond that, the picture is murky in nearly every case, with one exciting new exception. Of the top 35, I count 6—but this is most likely an undercounting:

— R1b-U106 > Y5051 (/FGC8564), the Ivanhoe cluster [Germanic/Iberian]
— R1b-PF7558 > A11720
— R1b-U152 > L408
— R1b-U152 > BY187600 [Italian/German/French]
— I2-M423 > Y23115 [Greek/Slavic]
— J2b-L283 > Y33795, the J2b Kohen cluster [Etruscan/Italian]

— [B]R1b-U106 > Y5051 (/FGC8564), the Ivanhoe cluster
The largest clearly European-introgressed Ashkenazi Y branch (~2.9%), R1b-Y5051 remains a puzzle despite investigators' best efforts. Its ultimate origin is clearly Germanic, and directly upstream of it within a TMRCA of 1500 ybp are English and Swedish individuals; within a TMRCA timescale of 1950 ybp, you get more Englishmen, plus a German from Baden (just shy of the Rhineland). However, right after this split from the English and Swedish branch, Y5051 splits into a Spanish/Portuguese branch and an Ashkenazi branch (R1b-FGC8564). Note, there are no known Sephardim in this branch, despite the putative Iberian connection.

Does this structure imply a Germanic > Iberian > Ashkenazi introgression? Despite theories about a British/English connection, the date perfectly fits a Visigothic arrival in Iberia. But is that at all connected to the lineage's presence in Ashkenazim? For now I think it's too early to say; if bona fide Sephardim turn up, we can assume so, but if closer German matches turn up, I'm inclined to chalk this up to a Rhenish convert. A problem for the latter theory is this branch's TMRCA of 500 CE (that is, it begins to rapidly split almost simultaneously with its split from the Iberian branch, and in turn with its split from the English-Swedish branch). This date isn't anomalous for an Ashkenazi founding line TMRCA (Q-Y2200, J2a-Y15223, and J1-Y5400, with its own idiosyncrasies, are in close range of it)—but it suggests that its incorporation into the Jewish population and its subsequent branching would have begun hundreds of years before the founding of the proto-Ashkenazi community in the Rhineland.

I'd love to hear more from R-U106 experts on this!

— R1b-PF7558 > A11720
R1b-A11720 (~1.7%) is also a difficult one to pin down, but by contrast, it's on a relatively understudied branch of R1b, and its nearest known sister branch shares with it a TMRCA of 4600 ybp. The branch itself has a TMRCA of 650 CE—relatively typical, but of little help in establishing the date of introgression. This sister branch, R1b-BY16680, seems to have a pan-Balkan distribution, with representation in Romania, Bulgaria, Serbia, and Kosovo/Albania; the FTDNA Haplotree also identifies two Italians and a Spaniard. Cousin branches, all falling under R-PF7563 (TMRCA: 4600 ybp) seem to be rooted in the Southern Balkans and Italy, with downstream presences in Lebanon and Saudi Arabia. This raises the possibility of some pre-exilic SE European introgression into the Ashkenazi gene pool, but without a closer Near Eastern link, I am inclined to chalk this up to introgression within Europe—either Italy or the Balkans. In any event, it is ultimately a European one.

— R1b-U152 > L408
Unfortunately, I can't give a TMRCA for this one, as YFull's U152 downstream structure is way too spare. A sister branch is found in an Italian line from Marche and a German line with deep roots in colonial America, whose surname is commonest in Baden-Württemberg and the Rhineland. One step further out are Italians from several regions, a French-Canadian, a Francophone Swiss man, as well as several English and English-origin Irish people. Someone who knows R-U152 > R-L2 better than I do (R. Rocca, cough) can comment, but it seems that despite an Italic origin, this broad grouping is pretty widely distributed within northwestern Europe. So an origin in Italy or Germany (or France) is possible; British seems historically untenable.

— R1b-U152 > BY187600
Similar issues as above. Parallel lines in Germany, England/Scotland(/Ireland), Luxembourg, Tatarstan, and Italy (Marche). No TMRCA calculation given. Origin in Italy or Germany (or France) is possible; British seems historically untenable.

— I2-M423 > Y23115
This one has been discussed before, and seems to be the best, most prominent candidate for Greek male introgression. Within a TMRCA of 2100 ybp, the branch has a Peloponnesian-Albanian sibling branch, and several steps up, it's mostly Greeks (with a Chuvash, a Russian, and a Ukrainian thrown in). The parent-branch's Greek history probably wasn't very deep before its entry into the pre-proto-Ashkenazi gene pool; its provenance is probably ultimately Old European > Balkan/Eastern European > Greek. AFAIK, the timeline is a bit early for a direct Slavic > Greek introgression, but northerly origins seem like a safe assumption. A caveat, though: there is a big gap between 100 BCE and the Ashkenazi branch's TMRCA of 1000 CE. It is theoretically possible that other non-Greek, non-Jewish Eastern European branches will show up under the immediate parent branch, I-A10959, leaving open the possibility of a Slavic introgression.

— J2b-L283 > Y33795, the J2b Kohen cluster
Despite a long history of wrangling over the origins of J2b-L283 as it appears in a small minority of Jews and Levantines, the recent Rome paper has gotten us close to a solution to this one. J2b-Y33975 falls directly under CTS6190 (TMRCA: 3100 ybp), which was found in an Iron Age Etruscan, and whose other descendants include Italians and Portuguese individuals. An ultimately Etruscan, and proximately Roman convert origin seems likely. The line's relatively early TMRCA of 200 CE is notable for two reasons: 1) it constrains the date of this conversion/introgression event to relatively early in Ashkenazi Jews' pre-Ashkenazi European history, and makes it likely that the line will show up in non-Ashkenazi Jews down the road, and 2) it resembles the early TMRCA's of other major priestly and Levite lines, suggesting that after the conversion event, this lineage's kohen status was preserved fervently, and most likely, across multiple communities.

EDIT: The Portuguese-flagged individual under sister-branch J2b-Y22038 (he shares it with a Tuscan, and an Italian with a surname mostly known from Liguria; TMRCA: 1900 ybp) is apparently a Sephardic Italian Jew of Portuguese origin, which implies, unless there is hidden structure under CTS6190 binding this group to the Ashkenazi group, that multiple CTS6190 individuals became Jewish in Roman-era Italy. I do not believe this individual identifies as a kohen.

One other line on another branch (J2b-Y36166), flagged as Portuguese, identifies as the descendant of an 11th-century Portuguese Jew, whose Caribbean Sephardic ancestor was adopted by an Anglo-Virginian in the early 19th century. There are too many questionable links in the story to fully credit it, but a Sephardic origin seems on the face of it likelier than an English one.

But wait, it gets stranger. The other Portuguese-flagged member of J2b-Y36166 (TMRCA: 225 ybp!) is [I]also an Anglo-Virginian; different surname, different part of the state, no known Sephardic, Italian, or Iberian roots. And under J-CTS6190*, yet a third Anglo-Virginian. What the hell?

Any thoughts on the two R1b-U152 lines? "Italy/France/Germany" is uncomfortably vague, and clearer resolution could provide genuine insights about early proto-Ashkenazi migration history.

Also: more incoming, re other lines. Following JewishDNA.net's list, I've compiled my own database of the top 50 Ashkenazi branches, which account for approximately 88% of Ashkenazi Y chromosomes. In addition to the above 6, 4 more are of definite European origin (though again, this is probably a modest underestimate).

To the extent that this is meaningful, when we discount the 1/8 of Ashkenazi Y-chromosomes not covered by the Top 50 branches, we find that no fewer than 20% of Ashkenazi branches are of European origin, accounting for no less than 10% of total Y chromosomes. Curiously, this male-line European contribution to the Ashkenazi population is extremely varied, split between Italian, Iberian, Greek, Balkan, Slavic, Germanic, and Atlantic Celtic lines.

Pylsteen
12-16-2019, 09:36 PM
I had calculated U152>L2>L408 to ca. 1100 AD (say 900-1300 AD) some time ago. Its subclade BY13145 (ca. 1400 AD) is mostly present in the Epstein clan of Frankfurt that dates back to this time.

Underneath U152>Z56 there are three Ashkenazi subclades, the most well-known L4 (ca. 300-600 AD), Y86521 (around the same age) and BY187600 (haven't calculated, but likely medieval as well).

Underneath U152>Z36 there is also a small Ashkenazi branch.

In most of these cases, these are medieval branches that come from clades dating back before 500 BC without any intermediate offshoots between 500BC-500AD. So until these appear, it remains difficult to be more precise than "Germany/France/(Northern) Italy". In my opinion, the Upper Rhine area/Lorraine might be an important source as opposed to Italy, but that's just a guess.

Principe
12-16-2019, 10:08 PM
Any thoughts on the two R1b-U152 lines? "Italy/France/Germany" is uncomfortably vague, and clearer resolution could provide genuine insights about early proto-Ashkenazi migration history.

Also: more incoming, re other lines. Following JewishDNA.net's list, I've compiled my own database of the top 50 Ashkenazi branches, which account for approximately 88% of Ashkenazi Y chromosomes. In addition to the above 6, 4 more are of definite European origin (though again, this is probably a modest underestimate).

To the extent that this is meaningful, when we discount the 1/8 of Ashkenazi Y-chromosomes not covered by the Top 50 branches, we find that no fewer than 20% of Ashkenazi branches are of European origin, accounting for less than 10% of total Y chromosomes. Curiously, this male-line European contribution to the Ashkenazi population is extremely varied, split between Italian, Iberian, Greek, Balkan, Slavic, Germanic, and Atlantic Celtic lines.

I am curious to see your top 50 list, any surprises in terms of frequencies of certain branches?

Looking forward to it Hartaisarlag!

hartaisarlag
12-18-2019, 12:02 AM
Some preliminary notes, and then some insights:

First, I should note, I am basing my frequency order on JewishDNA.net, just as Principe has in past posts on other threads. It's obvious that this is to be taken with a grain of salt. A representative, random sample would help us corroborate JewishDNA.net's survey (now a few years old) of FTDNA projects. Either way, I think the major branches are likely to hold up as the major branches, and the top 50 branches will still account for something like 7/8 of Ashkenazi Y chromosomes. But in combing through some of the small-to-medium branches, it's obvious that the reported relative frequencies are off. Moreover, besides #7 and 8 being tied in frequency on JewishDNA.net's list, from #12 onward, there are a good number of ties, sometimes multi-way. For the sake of convention, I'm still going to refer to them by their place in JewishDNA.net's frequency order.

The main improvement I'd say I've made over JewishDNA.net, descriptively, is cleaning up the terminal SNP terminology—in several cases, JewishDNA.net is insufficiently precise, and in some cases, overly precise. In addition, several clades show a split between Ashkenazi and Jewish non-Ashkenazi branches, which in most cases was not structurally clear when the JewishDNA.net description pages were compiled. Before YSEQ and the FTDNA Haplotree revealed the place of these Ashkenazi branches in the phylogeny, it was often not clear that Ashkenazi diversity was nested within wider Jewish diversity, in turn nested within wider European or Near Eastern diversity. In addition, JewishDNA.net rarely deals with interpretively meaningful TMRCA data, and has a bias toward narratives that see Iberia as an important venue for Ashkenazi ethnogenesis (i.e. if a branch has a Sephardic or Hispanic presence, it is inferred to have "passed through" Iberia, rather than ending up there, as well as Germany, via Italy or southern France).

Some things I'm interested in, in rough order of importance:

- Geographical/ethnic origin (Levantine? Northern West Asian? Mediterranean introgression? Continental European introgression?)
- Links to other Jewish groups (most often, Sephardic, but there are some scattered surprises)
- Patterns in the age of branches
- Unique regional patterns within the Ashkenazi world
- Patterns in the size and diversity of broader haplogroups
- Power laws in the share of total samples accounted for by the top branches



I'll start at the bottom of this list. See the graph below, which represents the share of sampled individuals accounted for by the top x branches:
35457

The top 5 branches comprise 36% of Ashkenazi males. The top 10 comprise 55%, the top 25 comprise 74%, and the top 50 comprise 88%.

Strikingly, the top 33 branches comprise both 20.5% of known Ashkenazi lines, and 79.8% of sampled Ashkenazi males. This is a perfect 80-20 Pareto distribution.



Among the top 50 branches, there are the following (total individual lines as a percentage of the 50 branches' 88.3% share, rather than out of 100%; whether this makes a difference is another question, though I suspect it does in the case of J1 and J2a, which are both very diverse among Ashkenazim):

9 J1 (14.0%)
9 R1b (9.6%) [6 of these lines are probably European, though one (R1b-A11720) has been disputed, and 3 are Near Eastern. ~23% of Ashkenazi R1b individuals belong to Near Eastern-origin lines.]
8 J2a (17.7%)
5 E-Z830 (12.9%)
5 G (10.6%)
3 R1a (9.5%) [2 of these lines are probably Near Eastern in some form, and 1 is Eastern European. ~94% of Ashkenazi R1a individuals belong to Near Eastern-origin lines, whatever the specific origin story of R1a-Y2619.]
3 I2 (2.2%) [2 of these lines are almost certainly European, and 1 (I2-BY2807) seems to be northern West Asian.]
2 J2b (1.3%) [1 of these lines is likely Italian (J2b-Y33795), and 1 is most likely Near Eastern (J2b-Z42983).]
2 E-M78 (1.8%)
1 Q (6.4%)
1 E1a (0.7%)
1 T (0.7%)
1 R2 (0.9%)



If anyone would like to chime in before I go on to discuss TMRCAs, highlight individual lines with interesting origin stories, and summarize known Jewish non-Ashkenazi connections, here's your chance.

StillWater
12-18-2019, 02:13 AM
Some preliminary notes, and then some insights:

First, I should note, I am basing my frequency order on JewishDNA.net, just as Principe has in past posts on other threads. It's obvious that this is to be taken with a grain of salt. A representative, random sample would help us corroborate JewishDNA.net's survey (now a few years old) of FTDNA projects. Either way, I think the major branches are likely to hold up as the major branches, and the top 50 branches will still account for something like 7/8 of Ashkenazi Y chromosomes. But in combing through some of the small-to-medium branches, it's obvious that the reported relative frequencies are off. Moreover, besides #7 and 8 being tied in frequency on JewishDNA.net's list, from #12 onward, there are a good number of ties, sometimes multi-way. For the sake of convention, I'm still going to refer to them by their place in JewishDNA.net's frequency order.

The main improvement I'd say I've made over JewishDNA.net, descriptively, is cleaning up the terminal SNP terminology—in several cases, JewishDNA.net is insufficiently precise, and in some cases, overly precise. In addition, several clades show a split between Ashkenazi and Jewish non-Ashkenazi branches, which in most cases was not structurally clear when the JewishDNA.net description pages were compiled. Before YSEQ and the FTDNA Haplotree revealed the place of these Ashkenazi branches in the phylogeny, it was often not clear that Ashkenazi diversity was nested within wider Jewish diversity, in turn nested within wider European or Near Eastern diversity. In addition, JewishDNA.net rarely deals with interpretively meaningful TMRCA data, and has a bias toward narratives that see Iberia as an important venue for Ashkenazi ethnogenesis (i.e. if a branch has a Sephardic or Hispanic presence, it is inferred to have "passed through" Iberia, rather than ending up there, as well as Germany, via Italy or southern France).

Some things I'm interested in, in rough order of importance:

- Geographical/ethnic origin (Levantine? Northern West Asian? Mediterranean introgression? Continental European introgression?)
- Links to other Jewish groups (most often, Sephardic, but there are some scattered surprises)
- Patterns in the age of branches
- Unique regional patterns within the Ashkenazi world
- Patterns in the size and diversity of broader haplogroups
- Power laws in the share of total samples accounted for by the top branches



I'll start at the bottom of this list. See the graph below, which represents the share of sampled individuals accounted for by the top x branches:
35457

The top 5 branches comprise 36% of Ashkenazi males. The top 10 comprise 55%, the top 25 comprise 74%, and the top 50 comprise 88%.

Strikingly, the top 33 branches comprise both 20.5% of known Ashkenazi lines, and 79.8% of sampled Ashkenazi males. This is a perfect 80-20 Pareto distribution.



Among the top 50 branches, there are the following (total individual lines as a percentage of the 50 branches' 88.3% share, rather than out of 100%; whether this makes a difference is another question, though I suspect it does in the case of J1 and J2a, which are both very diverse among Ashkenazim):

9 J1 (14.0%)
9 R1b (9.6%) [6 of these lines are probably European, though one (R1b-A11720) has been disputed, and 3 are Near Eastern. ~23% of Ashkenazi R1b individuals belong to Near Eastern-origin lines.]
8 J2a (17.7%)
5 E-Z830 (12.9%)
5 G (10.6%)
3 R1a (9.5%) [2 of these lines are probably Near Eastern in some form, and 1 is Eastern European. ~94% of Ashkenazi R1a individuals belong to Near Eastern-origin lines, whatever the specific origin story of R1a-Y2619.]
3 I2 (2.2%) [2 of these lines are almost certainly European, and 1 (I2-BY2807) seems to be northern West Asian.]
2 J2b (1.3%) [1 of these lines is likely Italian (J2b-Y33795), and 1 is most likely Near Eastern (J2b-Z42983).]
2 E-M78 (1.8%)
1 Q (6.4%)
1 E1a (0.7%)
1 T (0.7%)
1 R2 (0.9%)



If anyone would like to chime in before I go on to discuss TMRCAs, highlight individual lines with interesting origin stories, and summarize known Jewish non-Ashkenazi connections, here's your chance.

Just going by YFull, the Jews with I2-BY2807 could easily be Israelite/Judean in origin. This is because I-Y16419 has a hit in Saudi Arabia (south of Israel) and the others are, as you've observed, north of Israel. Did it skip over the land of Israel on its travels through the Middle East? I do agree that before it became Jewish, it would've came from Northwest Asia.

hartaisarlag
12-18-2019, 02:35 AM
As for TMRCAs, I am relying on YFull's latest estimates.

The larger the branch, generally, the more trustworthy the TMRCA estimate—for obvious reasons. For many of the smaller branches in JewishDNA.net's top 50, only 2 or 3 individuals have been uploaded to YFull, which yields a good handful of post-1250 TMRCAs, when the actual figure is probably earlier. Setting these aside, with very few exceptions, nearly all the branches have TMRCAs between 350 and 1050 CE. To be clear, there are a good number of branches which do not appear clearly on YFull, so naturally, YFull does not offer TMRCAs for them.

Half of the top 10 have TMRCAs between 750 (this is E-Y6938, and it keeps oscillating between 750 and 800) and 900 CE, which indicates an origin within a few generations of the first documented Ashkenazi communities in the Rhineland and Alsace-Lorraine.

Only 4 of the top 50 branches have earlier TMRCAs:

- #6, J1-Y3088/Z18271, as we know, is a special case with a pre-classical TMRCA, even though its largest Ashkenazi sub-branch (if YFull is to be trusted) has a TMRCA of 400 CE.
- #29, J2b-Y33795 (downstream of J2b-L283), also a Kohen branch, has a TMRCA of 200 CE, which given its likely Italian origin, is mighty suggestive of a classical Roman conversion/rape/NPE.
[Given the TMRCA, I expect that Jewish non-Ashkenazim will show up in this branch. One sample, labeled with a Dutch flag on YFull, claims Sephardic descent, but shares a TMRCA of 1000 CE with an Ashkenazi.]
- #35, J1-ZS2597 (downstream of J1-Y18297) has a TMRCA of 50 CE.
[Given the TMRCA, I expect that Jewish non-Ashkenazim will show up in this branch. The branch contains an ungrouped Mexican individual and an ungrouped Ashkenazi. The branch's nearest sibling, from which it diverges ca. 900 BCE, contains several Palestinians.]
- #50, J1-ZS1686 (downstream of J1-YSC76) has a TMRCA of 250 CE.
[Given the TMRCA, I expect that Jewish non-Ashkenazim will show up in this branch. The branch also diverges from a single Mexican individual ca. 250 CE.]

hartaisarlag
12-18-2019, 07:29 AM
The 10 almost certainly European-origin branches with a frequency of 0.4% or more are as follows:

- #10, R1b-FGC8564 (often mislabeled R1b-Y5051, downstream of R1b-U106), with a TMRCA of 500 CE. The enigma of this branch has been discussed before, and I go into greater detail earlier in this thread. The branch is Germanic in origin, but probably entered a Western Jewish source population in Spain. However, not a single Jewish non-Ashkenazi has been found in the branch. Given the TMRCA (which should be interpreted as a terminus ante quem for introgression), it seems likely that Jewish non-Ashkenazim will appear in the branch—otherwise, such an early common Ashkenazi ancestor, whether in Spain or in the Germanic lands, makes little sense.

- #12, R1b-A11720 (downstream of R1b-PF7558), with a TMRCA of 650 CE. This likely represents introgression within Europe, from an Old Balkan, Greek, or even Italian source. Introgression into the Near East can't be ruled out, but seems unparsimonious. The branch's lack of documented relatives more recent than 2600 BCE make it difficult to test any particular hypothesis, and leave open many possible dates for the introgression event.

- #20, R1b-L408 (downstream of R1b-U152), with a TMRCA of ca. 1100 CE (per Pylsteen). Origin in the Rhineland, France, or Northern Italy.

- #23, R1b-BY187600 (downstream of R1b-U152), TMRCA unknown. Origin in the Rhineland, France, or Northern Italy.

- #27, I2-Y23115 (downstream of I2-L621), with a TMRCA of 1000 CE. As discussed earlier, the source of this branch in the Ashkenazi community is either Slavic or Greek, though structural clues upstream suggest that the most proximate connection is Greek. The later TMRCA lends itself to two competing hypotheses: 1) late introgression into a pre-Ashkenazic Eastern European/Black Sea Jewish population, via either a Greek or Slavic source, or 2) earlier introgression into a Mediterranean Jewish source population via a Greek source, with Jewish non-Ashkenazi branch members out there, but yet to be documented.

- #29, J2b-Y33795 (downstream of J2b-L283), with a TMRCA of 200 CE. aDNA makes a very strong case for a classical-era Italian origin. Though how an Italian lineage gained priestly status—conversion, rape, or an NPE—will likely remain a mystery. Given the early TMRCA, I expect that Sephardim, Italkim, and North African Jews will eventually show up under the branch.

- #40, I2-Y52940 (downstream of I2-P78), with a TMRCA of 950 CE. This is a small branch. A note of caution: the Eastern European flags in its sister branch (and ungrouped under their parent branch) leave open the possibility that all of I-Y11261, with a TMRCA of 200 BCE, is Ashkenazi. However, this defies common sense. Assuming we are talking about a lineage with a TMRCA of 950 CE, whose closest cousins are Slavic (with a Romanian and a Swede in the mix too), we might be able to attribute this branch to early Slavic introgression in the pre-Ashkenazic Jewish communities of Eastern Europe. This date is just on the edge of the conventional narrative about the Jewish peopling of Eastern Europe, but it cannot be easily written off.

- #41, R1a-YP4848 (downstream of R-M458), a small branch with a TMRCA of 750 CE, might offer a parallel to I2-Y52940 - albeit with an even earlier Slavic introgression story. This date would pose some challenges for conventional Ashkenazi origin narratives. However, YFull does not provide a TMRCA for the branch’s primary sub-branch, R-M12402, which might turn out to be the Ashkenazi branch-defining SNP. Indeed, FTDNA records a non-Jewish Bulgarian under YP4848*, which parallels the presence of a Bulgarian in a sister-branch at YFull. This leaves open the distinct possibility of a more recent introgression.

- #42, R1b-Z18106 (downstream of L21), a small branch with no computed TMRCA, is nested within an overwhelmingly British Isles clade structure. I have no idea how it entered the Ashkenazi population, or when. Unlike “Ivanhoe”, it is Atlantic Celtic, rather than Germanic, in its deep origins, which would seem to rule out the possibility of a Germanic introgression in Visigothic Spain.

- #46, R1b-V3476 (downstream of R1b-DF27), with a TMRCA of 850 CE, a small branch. Structurally, this is an interesting one—as discussed earlier. 2 of its branches are Ashkenazi, and 1 Sephardic, but the ultimate origin of the branch is clearly Iberia. 850 CE represents a terminus ante quem for introgression, but no matter the date, Iberia is almost certainly the theater for it. The fact that multiple sub-branches of this branch became Ashkenazi suggest to me that the TMRCA of R1b-V3476 might have lived somewhere adjacent to both Iberia and Germany, such as southern France, and that its Sephardic sub-branch might represent a back-migration. Either way, the Iberian axis of this story is an interesting one.

As always, appreciate everyone’s thoughts about these branches. Let me know if you think any other branches that are commoner than say 0.4% are of European origin. I’m sure some are, but it’s too early to say in all other cases.

Pylsteen
12-18-2019, 09:50 AM
I had a quick look at R1b-BY187600; if I compare the BigY's I get an age of ~1422 AD (+/-249y) (assuming fixed rate of 144y/SNP for the Y500).
the sapp tool also gives an age of 1300-1450 AD.
So they - at least those who did this test - seem more recent than I thought. I also don't see any obvious matches for now that might push the age of this branch much further back.

There is a possible match to a non-Jewish German family.

hartaisarlag
12-18-2019, 04:37 PM
I had a quick look at R1b-BY187600; if I compare the BigY's I get an age of ~1422 AD (+/-249y) (assuming fixed rate of 144y/SNP for the Y500).
the sapp tool also gives an age of 1300-1450 AD.
So they - at least those who did this test - seem more recent than I thought. I also don't see any obvious matches for now that might push the age of this branch much further back.

There is a possible match to a non-Jewish German family.

Very interesting. How do this method’s estimates tend to compare to YFull’s estimates? Just want to make sure I’m not comparing apples to oranges, especially in the case of some of these other apparent European introgressions (and especially the smaller branches, whose YFull estimates are based on 2-4 samples).

Pylsteen
12-18-2019, 05:08 PM
Very interesting. How do this method’s estimates tend to compare to YFull’s estimates? Just want to make sure I’m not comparing apples to oranges, especially in the case of some of these other apparent European introgressions (and especially the smaller branches, whose YFull estimates are based on 2-4 samples).

I had L4 (incl. some genealogical information) at ~501 AD +/- few hundred years (seems same at Yfull) and Y86521 at ~402 AD +/- few hundred years. This branch only has one member at yfull which interestingly has TMRCA with a Sardinia sample at ca. 395 AD.

It only gives an approximate year though, with only a few samples, the number of mutations can differ quite a bit, and it doesn't take into account things such as amount of coverage etc.

I also have to find out yet what works best with the new Y700 results and groups with mixed Y500/Y700 results.

eolien
12-19-2019, 12:37 PM
- #46, R1b-V3476 (downstream of R1b-DF27), with a TMRCA of 850 CE, a small branch. Structurally, this is an interesting one—as discussed earlier. 2 of its branches are Ashkenazi, and 1 Sephardic, but the ultimate origin of the branch is clearly Iberia. 850 CE represents a terminus ante quem for introgression, but no matter the date, Iberia is almost certainly the theater for it. The fact that multiple sub-branches of this branch became Ashkenazi suggest to me that the TMRCA of R1b-V3476 might have lived somewhere adjacent to both Iberia and Germany, such as southern France, and that its Sephardic sub-branch might represent a back-migration. Either way, the Iberian axis of this story is an interesting one.



https://i.postimg.cc/hPzcBZhK/Capture5.png (https://postimages.org/)

Is the turkish guy the Sephardic one? And could you contact the Livorno one and confirm that he is of jewish origin? together with the 2 portugese/brasilian samples and other portugese samples, it looks like it is of Portugese origin? Of course during reconquista many people moved from France. Perhaps the surnames of the livorno and turkish guys could be informative.

Tz85
12-19-2019, 12:57 PM
Just wanted to throw this out.....the sub-clades of E-M78 are the real Israelites!......lol

Don't take everything I say seriously..

Claudio
12-19-2019, 06:13 PM
I just saw that on YFull myself, and I agree - it makes more sense that an Ashkenazi Jew perhaps made its way to Iraq - perhaps even during Ottoman times.

It’s possible,Nathan of Gaza for instance was born in Jerusalem yet his father was a German Jew:
35508

Claudio
12-19-2019, 06:40 PM
Any thoughts on the two R1b-U152 lines? "Italy/France/Germany" is uncomfortably vague, and clearer resolution could provide genuine insights about early proto-Ashkenazi migration history.

Also: more incoming, re other lines. Following JewishDNA.net's list, I've compiled my own database of the top 50 Ashkenazi branches, which account for approximately 88% of Ashkenazi Y chromosomes. In addition to the above 6, 4 more are of definite European origin (though again, this is probably a modest underestimate).

To the extent that this is meaningful, when we discount the 1/8 of Ashkenazi Y-chromosomes not covered by the Top 50 branches, we find that no fewer than 20% of Ashkenazi branches are of European origin, accounting for no less than 10% of total Y chromosomes. Curiously, this male-line European contribution to the Ashkenazi population is extremely varied, split between Italian, Iberian, Greek, Balkan, Slavic, Germanic, and Atlantic Celtic lines.

I read this the other day:
35509

Is the any evidence of this in the Y-dna of Sephardim and Ashkenazim?
In regards to sharing Y-dna subclades with Iran & Iraqi Jews?

hartaisarlag
12-19-2019, 07:35 PM
I read this the other day:
35509

Is the any evidence of this in the Y-dna of Sephardim and Ashkenazim?
In regards to sharing Y-dna subclades with Iran & Iraqi Jews?

So far, there is only one clear case of this in the top 50, and it is R2a-FGC13184. It's an obvious case of Mesopotamia/Iran > Europe (probably Eastern).

hartaisarlag
12-19-2019, 07:51 PM
https://i.postimg.cc/hPzcBZhK/Capture5.png (https://postimages.org/)

Is the turkish guy the Sephardic one? And could you contact the Livorno one and confirm that he is of jewish origin? together with the 2 portugese/brasilian samples and other portugese samples, it looks like it is of Portugese origin? Of course during reconquista many people moved from France. Perhaps the surnames of the livorno and turkish guys could be informative.

Confirmed members of this branch:
- multiple Turkish/Bulgarian Sephardic lines with variations on the surname Souroujon - great resource here: https://www.farhi.org/Documents/Souroujon.htm
- a Mexican named Villareal
- (upstream of these two), I am just now remembering that I found this Italian Jew among the top Family Finder matches of a Tunisian Jewish member of my branch - surname Soria from Livorno, apparently uploaded as part of the Avotaynu Western Sephardic project (still under lock and key...) EDIT: apparently a place name in old Castile, source of the well-known Soriano

I think the clustering of a Mexican w/a Turkish representative of the Souroujon family pretty conclusively eliminates the Romaniote theory. This is an Iberian-origin branch; R-Y31658 is Sephardic, and R-FGC20765 and R-FGC20816 are Ashkenazi. This 1 vs. 2 split is interesting to me. J-FGC5215, downstream of J1-Z640, is a case of the opposite.

There seems to be a Portuguese bent to the group, but there are plenty of Spanish and Mexican individuals not too far upstream. I don't know enough about classical-era Iberian Y-branch regional differentiation patterns, but it seems too early to make a specific conclusion about the Portuguese vs. Spanish origin of the branch.

Targum
12-19-2019, 10:54 PM
Confirmed members of this branch:
- multiple Turkish/Bulgarian Sephardic lines with variations on the surname Souroujon - great resource here: https://www.farhi.org/Documents/Souroujon.htm
- a Mexican named Villareal
- (upstream of these two), I am just now remembering that I found this Italian Jew among the top Family Finder matches of a Tunisian Jewish member of my branch - surname Soria from Livorno, apparently uploaded as part of the Avotaynu Western Sephardic project (still under lock and key...) EDIT: apparently a place name in old Castile, source of the well-known Soriano

I think the clustering of a Mexican w/a Turkish representative of the Souroujon family pretty conclusively eliminates the Romaniote theory. This is an Iberian-origin branch; R-Y31658 is Sephardic, and R-FGC20765 and R-FGC20816 are Ashkenazi. This 1 vs. 2 split is interesting to me. J-FGC5215, downstream of J1-Z640, is a case of the opposite.

There seems to be a Portuguese bent to the group, but there are plenty of Spanish and Mexican individuals not too far upstream. I don't know enough about classical-era Iberian Y-branch regional differentiation patterns, but it seems too early to make a specific conclusion about the Portuguese vs. Spanish origin of the branch.

IMHO Soriano & Sourojon is the same family name ; simply language-shifted; this affects all Jews, since names are being transcribed and re-transcribed in Hebrew, Latin, Arabic, Cyrillic & Greek alphabets.

StillWater
12-20-2019, 12:10 AM
So far, there is only one clear case of this in the top 50, and it is R2a-FGC13184. It's an obvious case of Mesopotamia/Iran > Europe (probably Eastern).

I recall posting 2 examples in the Aegean thread., maybe one of them wasn't in the top 50.

hartaisarlag
12-20-2019, 12:22 AM
IMHO Soriano & Sourojon is the same family name ; simply language-shifted; this affects all Jews, since names are being transcribed and re-transcribed in Hebrew, Latin, Arabic, Cyrillic & Greek alphabets.

Is there any evidence for this? The TMRCA is a full 900 years ago.

hartaisarlag
12-20-2019, 12:23 AM
I recall posting 2 examples in the Aegean thread., maybe one of them wasn't in the top 50.

Which are these? Could’ve missed them.

StillWater
12-20-2019, 12:39 AM
Which are these? Could’ve missed them.

This was one of them: https://yfull.com/tree/R-Y2632/ However, I'm not stating that it's originally Mesopotamian in origin - simply that it at least passed through there.

hartaisarlag
12-20-2019, 01:35 AM
This was one of them: https://yfull.com/tree/R-Y2632/ However, I'm not stating that it's originally Mesopotamian in origin - simply that it at least passed through there.

That one’s plain weird - it just makes the Top 50. Still not clear to me based on even the FTDNA phylogeny which branches are Ashkenazi and which aren’t.

Claudio
12-20-2019, 05:51 AM
So far, there is only one clear case of this in the top 50, and it is R2a-FGC13184. It's an obvious case of Mesopotamia/Iran > Europe (probably Eastern).

Is that regarding just Ashkenazim or Sephardim also?

hartaisarlag
12-20-2019, 05:56 AM
Is that regarding just Ashkenazim or Sephardim also?

I have not seen any Sephardim linked to this branch.

eolien
12-20-2019, 06:56 AM
Is there any evidence for this? The TMRCA is a full 900 years ago.

I think they cannot be the same name for linguistic reasons. Moreover the Livorno guy is probably descendant of famous (de) Soria banker family who can trace its origin to 16th century to Livorno. That makes me think that they are of Converso origin who often took (or retook) real or imaginary city names as surnames which would not have arisen suspicion in Iberia.

Claudio
12-20-2019, 10:05 AM
I have not seen any Sephardim linked to this branch.

But separetly to Ashkenazim.. do you know if Sephardim share Ydna with Iraq & Iranian Jews at any kind of elevated rate?
as Article snippet I originally posted would maybe be suggestive of migrations to Spain & North Africa?

Targum
12-20-2019, 11:56 AM
Is there any evidence for this? The TMRCA is a full 900 years ago.

Evidence is the language shift; see spellings/pronunciations if the names כהן &לוי for example

hartaisarlag
12-20-2019, 12:01 PM
Evidence is the language shift; see spellings/pronunciations if the names כהן &לוי for example

I wouldn’t call that evidence. Unless there’s a 900-year tradition of origins in Soria, ES, I think it’s a superficial resemblance. Not to mention, the Soujourouns clusters with a Mexican to the exclusion of Soria.

hartaisarlag
12-20-2019, 12:08 PM
But separetly to Ashkenazim.. do you know if Sephardim share Ydna with Iraq & Iranian Jews at any kind of elevated rate?
as Article snippet I originally posted would maybe be suggestive of migrations to Spain & North Africa?

Worth looking into, but not that I’m aware of. Unfortunately, no other Jewish community’s specific Y-lineages are nearly as well documented as Ashkenazi ones.

Targum
12-20-2019, 04:57 PM
I wouldn’t call that evidence. Unless there’s a 900-year tradition of origins in Soria, ES, I think it’s a superficial resemblance. Not to mention, the Soujourouns clusters with a Mexican to the exclusion of Soria.

Hebrew and linguistics background here: respectfully disagree brother

hartaisarlag
12-20-2019, 05:45 PM
Hebrew and linguistics background here: respectfully disagree brother

I have some linguistics background as well, and speak Hebrew, plus, there’s nothing Hebrew about those names. Without a solid, well-articulated etymological link, I can’t assume there’s a connection based on the presence of the Sor- prefix (my mistake on the spelling just above).

Claudio
12-21-2019, 11:16 AM
Worth looking into, but not that I’m aware of. Unfortunately, no other Jewish community’s specific Y-lineages are nearly as well documented as Ashkenazi ones.

Ok.
This may have been asked earlier in thread (and I’ve missed it) but how many or what percentage of Ashkenazi Ydna is directly shared with Sephardim or at least branched off from an older common pre Sephardim pre Ashkenazim Ydna ancestor?

hartaisarlag
12-21-2019, 08:16 PM
Ok.
This may have been asked earlier in thread (and I’ve missed it) but how many or what percentage of Ashkenazi Ydna is directly shared with Sephardim or at least branched off from an older common pre Sephardim pre Ashkenazim Ydna ancestor?

Good question - I will probably post a full run-down soon. While I haven’t calculated the figure of total % of individual lines, at this point, *at most* only 30% of Ashkenazi Top 50 branches have shown Sephardic connections - and only about half of these are ironclad, structurally sound connections (in which an Ashkenazi-Sephardi split can be explicitly identified and dated).

I expect the true figure is much higher, but my job here is to document what we definitely know. And considering we knew virtually none of this 5 years ago, it’s pretty awesome to be able to say, ‘such and such Ashkenazi branch and such and such Sephardi branch have a common ancestor ca. 800 CE’.

SUPREEEEEME
12-21-2019, 08:31 PM
Ok.
This may have been asked earlier in thread (and I’ve missed it) but how many or what percentage of Ashkenazi Ydna is directly shared with Sephardim or at least branched off from an older common pre Sephardim pre Ashkenazim Ydna ancestor?

My subclade has both Ashkenazi and Sephardi Jews (and others). I can't speak for other subclades, though.

hartaisarlag
12-21-2019, 08:34 PM
My subclade has both Ashkenazi and Sephardi Jews (and others). I can't speak for other subclades, though.

Which?

SUPREEEEEME
12-21-2019, 09:01 PM
Which?

J-L70. It's also been found in a Romaniote Jew, Djerban Cohanim (M318 subclade), Libyan Jews (M318 subclade), Azeri Jews and Bukharan Jews.

On FTDNA Projects, I've noticed that it's mainly J-PF5456 (downstream of L70) that's found in both Ashkenazim and Sephardim - and not so much J-P244 (also downstream of L70) it would seem. There are a decent number of people from Spain, Portugal and the Americas with Sephardic surnames that belong to this clade. I have also encountered some on the Facebook J2 group.

I've also encountered an individual who believed his paternal line was Italian only to discover it was actually Jewish - but I'm not sure whether it would be Sephardi, Italqi etc...

hartaisarlag
12-21-2019, 09:22 PM
J-L70. It's also been found in a Romaniote Jew, Djerban Cohanim (M318 subclade), Libyan Jews (M318 subclade), Azeri Jews and Bukharan Jews.

On FTDNA Projects, I've noticed that it's mainly J-PF5456 (downstream of L70) that's found in both Ashkenazim and Sephardim - and not so much J-P244 (also downstream of L70) it would seem. There are a decent number of people from Spain, Portugal and the Americas with Sephardic surnames that belong to this clade. I have also encountered some on the Facebook J2 group.

I've also encountered an individual who believed his paternal line was Italian only to discover it was actually Jewish - but I'm not sure whether it would be Sephardi, Italqi etc...

Not that this isn't significant, but my investigation (for now) is focused on primarily-to-exclusively Ashkenazi branches with a Common Era TMRCA, and in my upcoming post, am only reporting on cases where their direct sibling-branches are also clearly of Jewish (but non-Ashkenazi) origin. J-L70 is too deep and diverse a branch to qualify in this sense. J-FGC21085, the larger of the two Ashkenazi L70 branches, does not appear to have a Sephardic sibling branch; J-Z39057's nearest relation (J-F801) is too mixed (and based on rough evidence, too distant, TMRCA-wise) to be considered good evidence of such a connection, even though it contains Hispanic people who might be of ultimately Sephardic origin.

Thanks for alerting me to the J-M318 connection; anything Tunisian + Libyan Jewish is of interest. Now, it's possible that the clade has a deep, pan-Jewish/Levantine affinity. However, there are too many intervening branches with no Jewish pedigree for this to be assumed fact.

StillWater
12-22-2019, 12:46 AM
I've been trying to guess what YDNA my other ancestors might be under. Though I can't find any with a probability I'd wager much money on, 2 of them might be J-Y36257 and J-PF5456. All I did was match surnames and country locations. Probably raised the chances from winning the lottery to a couple %, but it's the best I can do for now. I may get 23andMe just to search by surname and see my matches' YDNA. Any information on either J-Y36257 and J-PF5456? All I've found about J-Y36257 is that it's very isolated, and the closest branches have South Italians/Sicilians, North Caucasians, and Saudis.

SUPREEEEEME
12-22-2019, 07:53 AM
I've been trying to guess what YDNA my other ancestors might be under. Though I can't find any with a probability I'd wager much money on, 2 of them might be J-Y36257 and J-PF5456. All I did was match surnames and country locations. Probably raised the chances from winning the lottery to a couple %, but it's the best I can do for now. I may get 23andMe just to search by surname and see my matches' YDNA. Any information on either J-Y36257 and J-PF5456? All I've found about J-Y36257 is that it's very isolated, and the closest branches have South Italians/Sicilians, North Caucasians, and Saudis.

I was quite lucky to find my 6× maternal great grandfathers Y-DNA. I did so by scrolling through the Iberian Ashkenaz FTDNA group and happened to recognise the name.

Claudio
12-22-2019, 11:39 AM
Good question - I will probably post a full run-down soon. While I haven’t calculated the figure of total % of individual lines, at this point, *at most* only 30% of Ashkenazi Top 50 branches have shown Sephardic connections - and only about half of these are ironclad, structurally sound connections (in which an Ashkenazi-Sephardi split can be explicitly identified and dated).

I expect the true figure is much higher, but my job here is to document what we definitely know. And considering we knew virtually none of this 5 years ago, it’s pretty awesome to be able to say, ‘such and such Ashkenazi branch and such and such Sephardi branch have a common ancestor ca. 800 CE’.

Interesting..
I’ll look forward to the future post.
Ultimately,what I’m really interested to find out (and eventually separate) is what percentage of Ashkenazim Y-dna is shared with Sephardim from Sephardic Jews migrating directly to France/Germany prior to the formation of the Ashkenazi communities of the early Middle Ages and alternatively what percentage is actually Shared Sephardim/Ashkenazim Y-dna going back to perhaps a much older proto Western Jewish population? and/or directly to the East Mediterranean & Near East?

hartaisarlag
12-22-2019, 03:20 PM
Interesting..
I’ll look forward to the future post.
Ultimately,what I’m really interested to find out (and eventually separate) is what percentage of Ashkenazim Y-dna is shared with Sephardim from Sephardic Jews migrating directly to France/Germany prior to the formation of the Ashkenazi communities of the early Middle Ages and alternatively what percentage is actually Shared Sephardim/Ashkenazim Y-dna going back to perhaps a much older proto Western Jewish population? and/or directly to the East Mediterranean & Near East?

My biggest conjecture from the data, and right now it’s just very early conjecture, is that there are two types of divergence represented, one earlier and one later. However, in the case of late divergences (ca. 800-900 CE), a common source in Provence/Occitania might be more parsimonious than an ultimate origin among Jews in Spain.

At least one of the Top 50 (R-V3476, described just above) is incontrovertibly Iberian in origin, but the structure of the branch makes a single direct Spain > northern France/Germany migration unlikely. Another ultimately Levantine branch of J1 in the Top 50 has a structure that implies an actual Sephardic or proto-Sephardic origin prior to entering the Ashkenazi community.

Claudio
12-23-2019, 12:51 PM
My biggest conjecture from the data, and right now it’s just very early conjecture, is that there are two types of divergence represented, one earlier and one later. However, in the case of late divergences (ca. 800-900 CE), a common source in Provence/Occitania might be more parsimonious than an ultimate origin among Jews in Spain.

At least one of the Top 50 (R-V3476, described just above) is incontrovertibly Iberian in origin, but the structure of the branch makes a single direct Spain > northern France/Germany migration unlikely. Another ultimately Levantine branch of J1 in the Top 50 has a structure that implies an actual Sephardic or proto-Sephardic origin prior to entering the Ashkenazi community.

What’s your opinion on the North African Admixture in Ashkenazi Jews?
Obviously Sephardic Jews have much higher levels,which I imagine was added in Arab Iberia then more recent post expulsion 15th century onwards in North Africa,but even the Greek Jews of Ioaninna have elevated North African Admixture.
Correct me if I’m wrong,But for Ashkenazim to have there current level of North African admixture one would at first glance imagine that Ashkenazim are solely descended of Sephardim mixed with Minor North & East Euro admixture.
Being that we think this is not the case does this imply that the proto Western Jewish populations that seeded both Ashkenazim & Sephardim already had a certain level of North African Admixture prior to arriving in Provence,Iberia,Rhineland etc?
We know that a lot of Jews in imperial Roman italy came from Cyrene and Alexandria (especially Cyrene) we also Know that post Kitos war Cyrenaican diaspora that was expulsed basically was absorbed by imperial era Jewish populations of Antioch,Aegean Greece /Anatolia /syria,essentially what became the Byzantine diaspora populations.
So I’m thinking Italian and Greek based Proto Western Jewish populations that seeded Provence May have already had elevated North African admixture like the Ioaninna Greek Jewish samples?

hartaisarlag
12-23-2019, 09:21 PM
Next, Jewish non-Ashkenazi links.

First, it needs to be noted that only a small minority of the 50 largest Ashkenazi branches have been directly linked to Jewish non-Ashkenazi branches. Given what most of us agree re: the nature of the Western Jewish ethnogenesis, the largest factor here is likely undersampling of Jewish non-Ashkenazim. However, without direct evidence of such connections, we can't assume one exists for any given branch.

In the minority of cases where proof of a connection does exist, structure and dating can help give us some preliminary clues about the nature of shared roots between Ashkenazim and non-Ashkenazim (in most, but not all cases, Sephardim). These factors have generally not been taken into account by JewishDNA.net, and the presence of Iberian and Sephardic connections has been misconstrued to indicate that a Jewish branch "passed through" Iberia, rather than arriving in both Iberia and Ashkenaz via some indeterminate Mediterranean Jewish source. Unfortunately, it is hard to get more specific in most cases—but thoughtful speculation is possible and welcome.

The branches with connections to Jewish non-Ashkenazi communities are as follows. I should note that in a lot of cases, the connection is cinched using Hispanic or Iberian samples; in cases of reliable knowledge of a common Middle Eastern origin + a 1st millennium CE divergence, this seems like a reasonable proxy for a Sephardic connection. This is a necessity, given how many more Hispanics and Iberians have tested than Sephardim. It is worth noting that at most, a non-Ashkenazi connection can be claimed for 16 of the top 50 branches on the basis of genetic evidence. While this is surely an underestimate, it goes to show how far we have to go, and how many variables we have yet to solve for.

First of all, only 9 Ashkenazi branches of the Top 50 (if we count #52, it becomes 10) can be connected to either Jewish non-Ashkenazi or Hispanic branches in a structurally sound way. That is to say, we can identify and date a clear split between Ashkenazi and non-Ashkenazi branches. In some cases this takes the form of a bifurcation; in others, the shape is a bit more complicated. 

It is also worth noting that in some of the following cases, the split appears to be late classical (ca. 250-400 CE), but in others, later early medieval (800-900 CE). This distinction is potentially meaningful. My personal hypothesis: splits ca. 800-900 CE might reflect a mutual origin in southern France, namely, in the longstanding Jewish communities of Narbonne, Arles, and Marseille, which are known to have been connected to both the early Jewish communities of Spain/Catalonia and northern France/Rhineland. I cannot think of any other Jewish region that was connected by migration to both Spain and Germany at this date. Earlier splits are harder to locate, and will depend on additional samples.

#5, E1b-Y6938 - Sibling branch: Hispanic, Sephardic, North African Jewish (Split: 400 CE)

E1b-Y6938 (TMRCA: 750 CE, downstream of E1b-M34 > L791) splits ca. 400 CE from E1b-Y102667, which has been confirmed in a Libyan Jew, a Tunisian Jew of claimed Turkish Sephardic origin, and a Puerto Rican. It is also likely that another Libyan Jew, other Tunisian Jews, an Algerian Jew, a Mexican, a Peruvian, and a Sicilian (west, if you must know) belong to this branch.

The date of this split coincides with the last century of the Western Roman Empire, and indeed, all of this group’s members, including Tripolitanian Jews, have their origins in territories held by the Western Roman Empire in its latest phase. Beyond this, it is very hard to make conclusions about the site of these sister branches’ common origin (that is, E1b-Y6923’s place/community of origin). A lot hinges on whether its North African Jewish members can be understood as Sephardic or pre-Sephardic North African in origin. Is this an Ashkenazi-Sephardi branch, or something broader than that?

#6, J1-Y3088/Z18791 - SBs: Pan-Jewish

The J1-Y3088/Z18791 Kohanic line has been discussed at length — and while YFull is currently offering a TMRCA of 1100 BCE, those who know tend to prefer the Behar et al. estimate ca. 570 BCE. The internal structure on this line seems to be fairly jumbled in its geography, but in short, even its largest Ashkenazi branch (J1-Y5400/S12192, TMRCA: 400 CE) has Moroccan and Algerian Jewish representation. Other branches include Yemenite, Iraqi, Turkish, Lebanese, and Greek Jews, Italians, and Iberians/Hispanics.

This branch’s structure is sui generis and almost certainly reflects a deep (at least Persian period) tradition of Kohanic lineage; it might or might not be possible to generalize to Jewish migrational history at large from its phylogeography.

#9, J1-ZS2728 - SBs: Hispanic and Italian/Romaniote (Split: 850/900 CE)

J1-ZS2728 (TMRCA: 900 CE) usually gets described as J1-L816, but J1-L816 is too broad to be considered an Ashkenazi founding lineage. Rather, ca. 850 CE, J1-ZS2728 splits from J1-Y34527, found in several Mexicans/Spaniards with a fairly recent TMRCA, and curiously/more distantly, a German Jew.

This latter branch could theoretically be an example of an Ashkenaz > Sepharad migration or a Sepharad > Ashkenaz migration; I am inclined to believe the latter. I am also inclined to believe that such a late split might indicate a common Jewish origin in southern France for this branch.

Particularly because downstream of this split is another Ashkenazi—non-Ashkenazi split; several individuals not shown on YFull, Romaniote Jews with a name indicating a pre-Inquisition origin in Apulia (in addition to one unknown Spanish individual), split off from the rest of J-ZS2728, probably around 850 to 900 CE. Given the evidence suggesting a partial Apulian Jewish origin for the earliest Ashkenazim, it is highly tempting to bag this as evidence.

#19, J1-FGC5215 - SBs: Multiple Hispanic branches (Split: 350 CE)

Spotlighted in Waas et al’s “Haplogroup J-Z640 - Genetic Insight into the Levantine Bronze Age”, J1-FGC5215 (TMRCA: 1050 CE) is one of three descendants of J1-FGC5206 (TMRCA: 350 CE), the other 2 of which are likely Sephardic, with descendants in Spain, Mexico, and Chile. More than for any other Ashkenazi line with documented Sephardic sibling-branches, this split structure implies the strong possibility of a Spain > Germany migration, but a common origin in southern France or Italy cannot be ruled out.

#21, R1b-FGC20980 - SB: Hispanic, Sephardic (Split: 800 CE)

R1b-FGC21047 (downstream of R-V88) is the largest Ashkenazi R1b branch of West Asian/Levantine origin, as well as the largest Ashkenazi R1b branch with an established pedigree in the wider Jewish world. Though its YFull samples cluster closely together (I expect the addition of a Bulgarian Jewish line might push the 1725 CE TMRCA back, but it’s also worth noting that Bulgarian Jewry is mixed in its origins), it diverges ca. 800 CE from a Sephardic/Hispanic branch (R1b-Y34349) containing a Greek Jew from Thessaloniki, a Mexican, and a Peruvian. This branch itself has a TMRCA of 800 CE. Stuck upstream at FTDNA are a German Jew (assigned as a likely member of the Ashkenazi branch) and another Mexican (assigned as a likely member of the Sephardic/Hispanic branch.

The late split, coincident with that of J-ZS2728 from its Iberian counterpart, is interesting. Again, it seems to me that in these cases, a common origin in southern France is more parsimonious than simultaneous Italy > Spain and Italy > Germany migrations. I’d be curious to hear peoples’ thoughts.

#22, R2a-FGC13211 - SB: Mizrahi (Split: 850 CE)

R2a-FGC13211 (TMRCA: 1725 CE; split from sibling: 850 CE) is a unique case. Namely, it is the only top 50 branch that clearly represents a Mesopotamian/Iranian/Caucasian Jewish migration into Ashkenazi lands (and quite likely, Eastern Ashkenazi lands). The sibling branch is represented among several Iranian and Iraqi Jewish families, and the Mizrahi > Ashkenazi introgression could have happened any time between 850 and 1725 (I suspect the latter is way too late, and that the addition of more branch members would push it back). Another sibling branch appears to have introgressed from a Mizrahi source into the Iraqi and Kuwaiti population.

To my knowledge, this branch hasn’t been found in any western or central Ashkenazim, which offers an important hint about the route of introgression. Upstream are Iranians, Armenians, and Arabs, which makes determining the source population of the original Mizrahi branch difficult to ascertain.

#33, R1b-Y19847a - SB: Hispanic (Split: 350 CE)

R1b-Y19847a (TMRCA: 350 CE), downstream of R1b-Z2103, has a sibling (split: 350 CE) in R1b-Y14987*, found in a Chilean. While FTDNA groups this individual as a branch parallel to two main Ashkenazi branches, YFull seems to identify him as an upstream splitter. I’m not sure which account to trust, but a primary Sephardic/Iberian vs. Ashkenazi split has somewhat different implications than a 3-way split between 1 Sephardi and 2 Ashkenazi branches.

#46, R1b-V3476 - Multiple Ashkenazi branches; SB: Hispanic/Sephardic (Split: 850 CE)

R1b-V3476 (TMRCA: 850 CE), downstream of R1b-DF27 has already been discussed above—two of its branches are Ashkenazi, while one has been found in a Mexican, a Livornese Sephardic Jew, and an extended family of Turkish and Bulgarian Jews. This split structure makes a direct Spain > Germany migration rather unlikely, despite the ultimate Iberian non-Jewish origin of the line. A Spain > Southern France > [Spain; Germany] pattern is proposed.

#50, J1-ZS1686 - SB: Hispanic (Split: 250 CE)

J1-ZS1686 (TMRCA: 250 CE) appears to split ca. 250 CE from a Mexican.

I will tack on another branch that falls just outside the top 50, which was covered in an earlier post. I know the Tunisian Jew in the sibling branch, and find his case instructive (among other things, this is the only Ashkenazi branch I know of besides E-Y6938 and J1-Y3088 with a documented North African Jewish presence.

#52, R1b-FGC14600 - SB: North African Jewish/Hispanic (Split: 900 CE)

R-FGC14600 (TMRCA: 500 ybp), downstream of R-L584, splits ca. 900 CE from a branch including one Tunisian Jew and one Hispanic individual. The Tunisian Jew indicates that his surname line has close ties to Livorno, which matches the phylogenetic evidence of a Sephardic origin for his sub-branch. Notably, among Ashkenazim, this line is attested in old Rhineland communities, and is associated with the Shapiro family, suggesting a link to the early Ashkenazi community of Speyer.



For the following branches, there is circumstantial evidence of either a Sephardic or Italian Jewish connection, and if I were a betting man, I’d put money on the connection being genuine. That said, in these cases there is 1) no structural evidence of a split, or muddled evidence of such a split, and 2) no means by which to date a hypothetical split.

#2, J2a-L556 - Possible SB: Hispanic

J2a-L556 (TMRCA: 850 CE) has according to FTDNA been found a Mexican, a Chilean, and a Honduran (as well as 2 Italians and a Pakistani) as a terminal SNP.

#3, G2b-Y12975 - PSB: Italian Jew

G2b-Y12975 (TMRCA: 900 CE) has apparently been found in a single-SNP YSEQ test in an Italian Jew with deep roots in Tuscany, and prior to that, Bologna.

#4, Q1b-Y2200 - PSB: Greek/Hispanic/Italian

Q1b-Y2200 (/Q1b-Y2232) (TMRCA: 450 CE) splits into two main branches; one of them, Q1b-Y2197, has a primary split, likely earlier than 700 CE, between the Ashkenazi Q1b-Y2198, and the Greek/Italian/Portuguese/Brazilian Q1b-BZ72. This latter branch might be of Sephardic, Italki, or Romaniote provenance (and the age of Q1b-Y2200 implies a pre-Ashkenazi presence in the Western Jewish community, unless it’s a late classical non-Jewish introgression), but I’ve been able to find no information about these individuals.

#7, J2a-Y15223 - PSB: Hispanic, Egyptian Jew?

J2a-Y15223 (TMRCA: 550 CE) without downstreams has been confirmed in a Mexican, according to FTDNA, and estimated in an Egyptian Jew.

#13, J1-Y6384 - PSB: Turkish? Puerto Rican?

J1-Y6384 (TMRCA: 450 CE) is split twice vertically at FTDNA, but not at YSEQ. The culprits: a Turkish kit (Sephardic?) and upstream, a Puerto Rican kit.

#15, E1b-Y35934 - PSB: Italian Jew

E1b-Y35934 (TMRCA: 1250 CE) was found by Behar et al. 2017 in an Italian Jew—though without a pedigree, it’s impossible to say what community that Italian Jew belongs to.

#35, J1-ZS2597 - PSB: Hispanic

J1-ZS2597 (TMRCA: 50 CE) has two basal individuals (who apparently don’t cluster with each other)—a Mexican and a Ukrainian (Jew?).

An honorable mention goes to the notorious Ivanhoe branch:

#10, R1b-FGC8564 - SB: Hispanic

“Ivanhoe” has been discussed at length, but no resolution has been reached. The nearest branch to this Ashkenazi branch (TMRCA: 500 CE; Split: 500 CE) is Iberian, which would usually count as evidence of a Sephardic link if the branch’s ultimate provenance were Near Eastern. However, this branch’s ultimate provenance (in fact, immediately before splitting into pre-proto-Ashkenazi and Iberian (or possibly pre-proto-Sephardic) branches was Germanic. The possibility of a Germanic introgression outside of Spain cannot be ruled out, and without explicit proof of a Sephardic connection, this branch’s wider Jewish history remains an enigma.

hartaisarlag
12-23-2019, 09:24 PM
What’s your opinion on the North African Admixture in Ashkenazi Jews?
Obviously Sephardic Jews have much higher levels,which I imagine was added in Arab Iberia then more recent post expulsion 15th century onwards in North Africa,but even the Greek Jews of Ioaninna have elevated North African Admixture.
Correct me if I’m wrong,But for Ashkenazim to have there current level of North African admixture one would at first glance imagine that Ashkenazim are solely descended of Sephardim mixed with Minor North & East Euro admixture.
Being that we think this is not the case does this imply that the proto Western Jewish populations that seeded both Ashkenazim & Sephardim already had a certain level of North African Admixture prior to arriving in Provence,Iberia,Rhineland etc?
We know that a lot of Jews in imperial Roman italy came from Cyrene and Alexandria (especially Cyrene) we also Know that post Kitos war Cyrenaican diaspora that was expulsed basically was absorbed by imperial era Jewish populations of Antioch,Aegean Greece /Anatolia /syria,essentially what became the Byzantine diaspora populations.
So I’m thinking Italian and Greek based Proto Western Jewish populations that seeded Provence May have already had elevated North African admixture like the Ioaninna Greek Jewish samples?

This seems likely. So far, I don't think any evidence from the Top 50 Ashkenazi Y branches points to this, with the possible exception of a Libyan Jewish presence at the root of the non-Ashkenazi sub-branch (E-Y102667) of E-Y6923. I think a few more individuals on the tree will help us determine whether E-Y6923 arrived in Libya via Spain, or if it was already there. If the latter, it is worth considering Cyrene's Jewish community an original possible source of E-Y6923.

EDIT: I've been told that the Avotaynu project claims to have uniparental evidence for this trend, but they've resisted calls to share their data.

StillWater
12-23-2019, 10:26 PM
What’s your opinion on the North African Admixture in Ashkenazi Jews?
Obviously Sephardic Jews have much higher levels,which I imagine was added in Arab Iberia then more recent post expulsion 15th century onwards in North Africa,but even the Greek Jews of Ioaninna have elevated North African Admixture.
Correct me if I’m wrong,But for Ashkenazim to have there current level of North African admixture one would at first glance imagine that Ashkenazim are solely descended of Sephardim mixed with Minor North & East Euro admixture.
Being that we think this is not the case does this imply that the proto Western Jewish populations that seeded both Ashkenazim & Sephardim already had a certain level of North African Admixture prior to arriving in Provence,Iberia,Rhineland etc?
We know that a lot of Jews in imperial Roman italy came from Cyrene and Alexandria (especially Cyrene) we also Know that post Kitos war Cyrenaican diaspora that was expulsed basically was absorbed by imperial era Jewish populations of Antioch,Aegean Greece /Anatolia /syria,essentially what became the Byzantine diaspora populations.
So I’m thinking Italian and Greek based Proto Western Jewish populations that seeded Provence May have already had elevated North African admixture like the Ioaninna Greek Jewish samples?

You sure love this topic. U6a7a1b and L2a1 (I forget the rest) may be examples of North African admixture in Western Jews. The former used to be considered a good example, but there is now a Lebanese sample with it. Unless the Lebanese is Jewish, then it may be Judean. However, U6a7a1b still seems to be considered as Sephardi, despite existing among Ashkenazim. It could be that the Ashkenazim with it are of Sephardi origin.

jonahst
12-23-2019, 11:37 PM
#4, Q1b-Y2200 - PSB: Greek/Hispanic/Italian

Q1b-Y2200 (/Q1b-Y2232) (TMRCA: 450 CE) splits into two main branches; one of them, Q1b-Y2197, has a primary split, likely earlier than 700 CE, between the Ashkenazi Q1b-Y2198, and the Greek/Italian/Portuguese/Brazilian Q1b-BZ72. This latter branch might be of Sephardic, Italki, or Romaniote provenance (and the age of Q1b-Y2200 implies a pre-Ashkenazi presence in the Western Jewish community, unless it’s a late classical non-Jewish introgression), but I’ve been able to find no information about these individuals.


Awesome work! Where did you find the info on BZ72? And what are your thoughts on the fact that most of the parallel subclades under L245 (which, according to Yfull, branched off from each other around 4000 yars ago) are West Asian?

https://www.yfull.com/tree/Q-L245/

hartaisarlag
12-23-2019, 11:59 PM
Awesome work! Where did you find the info on BZ72? And what are your thoughts on the fact that most of the parallel subclades under L245 (which, according to Yfull, branched off from each other around 4000 yars ago) are West Asian?

https://www.yfull.com/tree/Q-L245/

Found it all on FTDNA's Haplotree; there are also an Italian and a Spanish BZ72 in "Phylogeography of human Y-chromosome haplogroup Q3-L275 from an academic/citizen science collaboration". Now that I've checked the Q-M242 project's results map (no results page is available publicly), I'm seeing two New Mexico locations—which makes me more inclined to believe there's a Sephardic link. The Greek location on the project's map is in Larissa, which was home to a Sephardic community. Haven't ever spent much time on this, but I'll e-mail the project admins for more on all this.

What's odd about this case is that a possibly Sephardic branch is nested high up within (one of the fairly early and deep examples of) an Ashkenazi branch. It's also still possible that there's nothing Sephardic about this branch at all, absent additional information.

I don't have any unique opinion on Q-L245 overall—but the parallel to certain R-Z93 branches seems instructive. A Central Asian introgression into the proto-Iranian population, followed by diffusion throughout the Near East, seems very possible.

jonahst
12-24-2019, 12:13 AM
Found it all on FTDNA's Haplotree; there are also an Italian and a Spanish BZ72 in "Phylogeography of human Y-chromosome haplogroup Q3-L275 from an academic/citizen science collaboration". Now that I've checked the Q-M242 project's results map (no results page is available publicly), I'm seeing two New Mexico locations—which makes me more inclined to believe there's a Sephardic link. The Greek location on the project's map is in Larissa, which was home to a Sephardic community. Haven't ever spent much time on this, but I'll e-mail the project admins for more on all this.

What's odd about this case is that a possibly Sephardic branch is nested high up within (one of the fairly early and deep examples of) an Ashkenazi branch. It's also still possible that there's nothing Sephardic about this branch at all, absent additional information.

I don't have any unique opinion on Q-L245 overall—but the parallel to certain R-Z93 branches seems instructive. A Central Asian introgression into the proto-Iranian population, followed by diffusion throughout the Near East, seems very possible.

There was also a guy of paternal Mexican ancestry on Anthrogencia who's Q-Y2197: https://anthrogenica.com/showthread.php?11166-Haplogroup-Q-M378-(Q1b1a)-Info&p=454500&viewfull=1#post454500

hartaisarlag
12-24-2019, 12:42 AM
There was also a guy of paternal Mexican ancestry on Anthrogencia who's Q-Y2197: https://anthrogenica.com/showthread.php?11166-Haplogroup-Q-M378-(Q1b1a)-Info&p=454500&viewfull=1#post454500

Would bet some fraction of my Hanukkah present that he falls under Q-BZ72. Good eye—I missed this!

Agamemnon
12-24-2019, 01:08 AM
#6, J1-Y3088/Z18791 - SBs: Pan-Jewish

The J1-Y3088/Z18791 Kohanic line has been discussed at length — and while YFull is currently offering a TMRCA of 1100 BCE, those who know tend to prefer the Behar et al. estimate ca. 570 BCE. The internal structure on this line seems to be fairly jumbled in its geography, but in short, even its largest Ashkenazi branch (J1-Y5400/S12192, TMRCA: 400 CE) has Moroccan and Algerian Jewish representation. Other branches include Yemenite, Iraqi, Turkish, Lebanese, and Greek Jews, Italians, and Iberians/Hispanics.

This branch’s structure is sui generis and almost certainly reflects a deep (at least Persian period) tradition of Kohanic lineage; it might or might not be possible to generalize to Jewish migrational history at large from its phylogeography.

Since we're talking about non-Ashkenazi links, there's a batch of Pontic Greeks from Gümüşhane (Argyropolis) and Erzurum (Theodosioupolis) under Z18271/Y3088 and whose closest matches are Ashkenazi, a few others are closer to a Romaniote Cohen, but most are specifically closer to Ashkenazim. Some could even potentially belong to sub-branches that seem to be ostensibly Ashkenazi, tellingly these sub-branches are xS12192 (which accounts for the bulk of Ashkenazi lineages). Another thing that is of interest is that the Ashkenazi clusters that are concerned here are mainly Litvak (for now).

This suggests that some of the sub-branches that are ostensibly Ashkenazi might not be specifically Ashkenazi, and that with the addition of samples such as these (with a finer analysis) that their MRCAs might be pushed back in time while Ashkenazim are relegated to narrow subclades. These individuals are very unlikely to be of Ashkenazi origin, it might be worthwhile to keep in mind that a partial origin of some of these sub-branches in Byzantine Jewry remains a possibility (despite the pan-Ashkenazi distribution of some of those sub-branches).

Johnny ola
12-24-2019, 01:21 AM
Since we're talking about non-Ashkenazi links, there's a batch of Pontic Greeks from Gümüşhane (Argyropolis) and Erzurum (Theodosioupolis) under Z18271/Y3088 and whose closest matches are Ashkenazi, a few others are closer to a Romaniote Cohen, but most are specifically closer to Ashkenazim. Some could even potentially belong to sub-branches that seem to be ostensibly Ashkenazi, tellingly these sub-branches are xS12192 (which accounts for the bulk of Ashkenazi lineages). Another thing that is of interest is that the Ashkenazi clusters that are concerned here are mainly Litvak (for now).

This suggests that some of the sub-branches that are ostensibly Ashkenazi might not be specifically Ashkenazi, and that with the addition of samples such as these (with a finer analysis) that their MRCAs might be pushed back in time while Ashkenazim are relegated to narrow subclades. These individuals are very unlikely to be of Ashkenazi origin, it might be worthwhile to keep in mind that a partial origin of some of these sub-branches in Byzantine Jewry remains a possibility (despite the pan-Ashkenazi distribution of some of those sub-branches).

My ancestry is from Gumushane but my ydna is G2a-L14 witch seems of Kura-Araxes origins.Can you mention Agamamemnon these ashkenazi related lineages from Gumushane and Eruzum?Keep in mind also that Eruzum was a hideway for Pontic Greeks during Geonocide period from Neo-Turk,Eruzum(Kars more specific) was a place under Russian army.My ancestors left from there to Greece,but my grandpa mention that the population there was not only Pontic Greeks but also Armenians.

Agamemnon
12-24-2019, 02:00 AM
My ancestry is from Gumushane but my ydna is G2a-L14 witch seems of Kura-Araxes origins.Can you mention Agamamemnon these ashkenazi related lineages from Gumushane and Eruzum?Keep in mind also that Eruzum was a hideway for Pontic Greeks during Geonocide period from Neo-Turk,Eruzum(Kars more specific) was a place under Russian army.My ancestors left from there to Greece,but my grandpa mention that the population there was not only Pontic Greeks but also Armenians.

The individuals' ancestors sought refuge in the Russian empire during the Pontic genocide and so their family names now have Russian endings, that being said this is no real obstacle. The names of the Z18271 individuals from Gümüşhane & Erzurum include Miridis (Μυρίδης), Ikonomidis (Οικονομίδης), Stoforidis (Στοφορίδης) and Iordanidis (Ιορδανίδης). One of the individuals is even predicted as ZS2434, which is completely Ashkenazi so far (it's also the branch I belong to).

While they are unlikely to be of Ashkenazi descent, these Pontians are undeniably of Jewish, and more specifically Kohanic (priestly) origin on their paternal line. This is a very peculiar case, because beyond small clusters in Italy, Iberia (mainly Portugal) and Yemen, Pontic Greeks are the only non-Jewish population where Z18271 really shows up, suggesting that Byzantine Jews were absorbed into the local population.

StillWater
12-24-2019, 02:20 AM
Since we're talking about non-Ashkenazi links, there's a batch of Pontic Greeks from Gümüşhane (Argyropolis) and Erzurum (Theodosioupolis) under Z18271/Y3088 and whose closest matches are Ashkenazi, a few others are closer to a Romaniote Cohen, but most are specifically closer to Ashkenazim. Some could even potentially belong to sub-branches that seem to be ostensibly Ashkenazi, tellingly these sub-branches are xS12192 (which accounts for the bulk of Ashkenazi lineages). Another thing that is of interest is that the Ashkenazi clusters that are concerned here are mainly Litvak (for now).

This suggests that some of the sub-branches that are ostensibly Ashkenazi might not be specifically Ashkenazi, and that with the addition of samples such as these (with a finer analysis) that their MRCAs might be pushed back in time while Ashkenazim are relegated to narrow subclades. These individuals are very unlikely to be of Ashkenazi origin, it might be worthwhile to keep in mind that a partial origin of some of these sub-branches in Byzantine Jewry remains a possibility (despite the pan-Ashkenazi distribution of some of those sub-branches).

Shit, you just found more evidence for Knaan Yavan. What's the distribution among Litvaks for this (Baltics vs Belarus vs regions closer to the enemy (Northeast Poland, Northern Ukraine etc)..)?

Claudio
12-24-2019, 10:34 AM
You sure love this topic. U6a7a1b and L2a1 (I forget the rest) may be examples of North African admixture in Western Jews. The former used to be considered a good example, but there is now a Lebanese sample with it. Unless the Lebanese is Jewish, then it may be Judean. However, U6a7a1b still seems to be considered as Sephardi, despite existing among Ashkenazim. It could be that the Ashkenazim with it are of Sephardi origin.

Yup! :thumb:
Cracking the origin of the North African Admixture in Ashkenazim,is the missing key to finally unlocking..
The Jew Vinci Code! :pop2:

hartaisarlag
12-24-2019, 05:34 PM
First, thanks for the additional commentary on J1-Z18271—I was inviting it. Did not know about that Pontic Greek cluster, but am pretty stunned. I'm fairly open to non-Italian/French/Rhenish origins for a decent share of the EEJ population, but so far this kind of investigation hasn't done much to cast those lines of origin into relief (especially if they're Near Eastern, rather than European additions). Hard not to agree with StillWater's impression here, though!

Meanwhile, on the topic of Jewish lines appearing in Italian and Portuguese gentiles, I've joined the Q group and gotten some speedy feedback from the admins. Under BZ72, they know of an Italian (probably Calabrian, per surname), a Portuguese, a second Portuguese (not in their project, and because to the flag on the Haplotree, I actually think Brazilian), two related New Mexicans, and a Greek Jew from Larissa with the last name Askenazi. I have also noticed two Italians in two otherwise all-Ashkenazi branches of Q-Y2232 (well, one is predicted Q-Y2198, and another is predicted Q-YP1004), which is not totally unheard of in Ashkenazi branches, but definitely demands investigation.

Some notes here:

- This could be coincidence, but the New Mexican family surname 1) sounds Jewish and 2) appears to be cognate with the Calabrian surname. Happy to share in DMs for anyone who'd like to weigh in.
- The Greek Jew named Askenazi poses a problem. In several Ashkenazi branches, I've run into Sephardim named Ashkenazi (or even Sarfati), and without fail, they've belonged to exclusively Ashkenazi sub-branches, validating their surname tradition. I've excluded these people from my analyses for obvious reasons. But, this Greek Jew, while nested within an otherwise Ashkenazi large branch, shares his sub-branch with only non-Ashkenazim (and falls downstream of some of them), whose roots are likelier Sephardic than anything else. This suggests to me an ultimately non-Ashkenazi origin for Mr. "Askenazi". Q-BZ72's terminus ad quem is 700 CE, given the age estimate of its parent-branch Q-Y2197, so maybe, just maybe, a more fitting surname might be "Pre-Proto-Ashkenazi".

hartaisarlag
12-24-2019, 06:11 PM
The individuals' ancestors sought refuge in the Russian empire during the Pontic genocide and so their family names now have Russian endings, that being said this is no real obstacle. The names of the Z18271 individuals from Gümüşhane & Erzurum include Miridis (Μυρίδης), Ikonomidis (Οικονομίδης), Stoforidis (Στοφορίδης) and Iordanidis (Ιορδανίδης). One of the individuals is even predicted as ZS2434, which is completely Ashkenazi so far (it's also the branch I belong to).

While they are unlikely to be of Ashkenazi descent, these Pontians are undeniably of Jewish, and more specifically Kohanic (priestly) origin on their paternal line. This is a very peculiar case, because beyond small clusters in Italy, Iberia (mainly Portugal) and Yemen, Pontic Greeks are the only non-Jewish population where Z18271 really shows up, suggesting that Byzantine Jews were absorbed into the local population.

Oh, and don't tell Elhaik. Who knows how he'll manage to fit this in...

John Doe
12-24-2019, 06:30 PM
Oh, and don't tell Elhaik. Who knows how he'll manage to fit this in...

He'll find a way don't worry ;-)

coffeeprince
12-24-2019, 07:20 PM
Hartaisarlag, I really appreciate the effort and work you've put into this. It's been a very enjoyable and enlightening read. On another note, I've check JewishDNA.net, and I don't even think my Y-DNA branch has found it's way on there yet.

hartaisarlag
12-24-2019, 07:59 PM
Hartaisarlag, I really appreciate the effort and work you've put into this. It's been a very enjoyable and enlightening read. On another note, I've check JewishDNA.net, and I don't even think my Y-DNA branch has found it's way on there yet.

My pleasure - I knew it’d spark good conversations, especially in cases where someone knows more about their own branch than I possibly could. Might keep posting, but from a 10,000-foot level (that is, about trends in groups of branches). A challenge that remains, for example, is connecting specific branches to different Levantine and West Asian populations. In a few cases, there’s actually something to say.

JewishDNA.net is pretty comprehensive AFAIK in its inclusion of various branches - but I’m not fully clear on their method for non-Ashkenazi branches. However, the labeling is often unclear. What’s your branch? Feel free to PM.

I fully expect we’ll have what to add by the end of 2020. Anyone who catches anything, please share it here!

Principe
12-24-2019, 11:42 PM
#3, G2b-Y12975 - PSB: Italian Jew

G2b-Y12975 (TMRCA: 900 CE) has apparently been found in a single-SNP YSEQ test in an Italian Jew with deep roots in Tuscany, and prior to that, Bologna.

#7, J2a-Y15223 - PSB: Hispanic, Egyptian Jew?

J2a-Y15223 (TMRCA: 550 CE) without downstreams has been confirmed in a Mexican, according to FTDNA, and estimated in an Egyptian Jew.

Just a quick note for the G-Y12975, there is evidence for a Provençal origin for at least part of the branch with what I believe are the Lattes and Franciscus lines, I think this branch is a strong candidate for a Y line that came out of the Southern French Jewish community into the Rhineland.

As for Y15223 I have quite a few Hispanic matches include a recent Y67 match at GD of 5. I will PM you all of the them!!

hartaisarlag
12-25-2019, 12:36 AM
Just a quick note for the G-Y12975, there is evidence for a Provençal origin for at least part of the branch with what I believe are the Lattes and Franciscus lines, I think this branch is a strong candidate for a Y line that came out of the Southern French Jewish community into the Rhineland.

As for Y15223 I have quite a few Hispanic matches include a recent Y67 match at GD of 5. I will PM you all of the them!!

Thanks for those matches! I'll look them over.

TBH I'm skeptical of the G-M377 project's designations—I think they're built on a lot of poorly-evidenced conjecture. The admin, who I like and who's quite talented, has become more cautious over the years, but has not revised these.

Franciscus is apparently a German or Swiss gentile line, and though the name is interesting, I don't think the "Tzarfati" label is anything more than a conjectural back-translation. I'd have to see more documentation for De Lattas/Leites, but that is a good find.

StillWater
12-25-2019, 01:15 AM
First, thanks for the additional commentary on J1-Z18271—I was inviting it. Did not know about that Pontic Greek cluster, but am pretty stunned. I'm fairly open to non-Italian/French/Rhenish origins for a decent share of the EEJ population, but so far this kind of investigation hasn't done much to cast those lines of origin into relief (especially if they're Near Eastern, rather than European additions). Hard not to agree with StillWater's impression here, though!

Meanwhile, on the topic of Jewish lines appearing in Italian and Portuguese gentiles, I've joined the Q group and gotten some speedy feedback from the admins. Under BZ72, they know of an Italian (probably Calabrian, per surname), a Portuguese, a second Portuguese (not in their project, and because to the flag on the Haplotree, I actually think Brazilian), two related New Mexicans, and a Greek Jew from Larissa with the last name Askenazi. I have also noticed two Italians in two otherwise all-Ashkenazi branches of Q-Y2232 (well, one is predicted Q-Y2198, and another is predicted Q-YP1004), which is not totally unheard of in Ashkenazi branches, but definitely demands investigation.

Some notes here:

- This could be coincidence, but the New Mexican family surname 1) sounds Jewish and 2) appears to be cognate with the Calabrian surname. Happy to share in DMs for anyone who'd like to weigh in.
- The Greek Jew named Askenazi poses a problem. In several Ashkenazi branches, I've run into Sephardim named Ashkenazi (or even Sarfati), and without fail, they've belonged to exclusively Ashkenazi sub-branches, validating their surname tradition. I've excluded these people from my analyses for obvious reasons. But, this Greek Jew, while nested within an otherwise Ashkenazi large branch, shares his sub-branch with only non-Ashkenazim (and falls downstream of some of them), whose roots are likelier Sephardic than anything else. This suggests to me an ultimately non-Ashkenazi origin for Mr. "Askenazi". Q-BZ72's terminus ad quem is 700 CE, given the age estimate of its parent-branch Q-Y2197, so maybe, just maybe, a more fitting surname might be "Pre-Proto-Ashkenazi".

How do you join DNA projects?

Also, as I said before, there is a video of a Syrian Jew surnamed Ashkenazi (I can't find it now). He's confident his surname comes from his ancestor doing business in Ashkenaz lands,as opposed to descending from an Ashkenazi. Davidi Weinberger mentions that there are Jews surnamed Yaponchik (haven't checked myself). Yaponchik means "little Japanese" - clearly an artificial surname.

There are also Ashkenazi Jews surnamed Ashkenazi - which is interesting.

hartaisarlag
12-25-2019, 01:35 AM
How do you join DNA projects?

Also, as I said before, there is a video of a Syrian Jew surnamed Ashkenazi (I can't find it now). He's confident his surname comes from his ancestor doing business in Ashkenaz lands,as opposed to descending from an Ashkenazi. Davidi Weinberger mentions that there are Jews surnamed Yaponchik (haven't checked myself). Yaponchik means "little Japanese" - clearly an artificial surname.

There are also Ashkenazi Jews surnamed Ashkenazi - which is interesting.

That's an interesting workaround. Hadn't heard of that before, but I could see it!

And most projects I'm interested in at FTDNA have allowed me to join as a guest—except the J-ZS241 one, which expelled me angrily when it found out I was "doing research". I actually haven't tested my Y chromosome at FTDNA, but I transferred my autosomal results from 23andMe to FTDNA's Family Finder several years back.

Shamash
12-25-2019, 11:09 AM
In my opinion a substantial number of Oriental Jews will turn out FGC3723+ which happens to be my J1-subclade.

As far as Yemen is concerned there are only three Jewish J1 samples (all three FGC3723+):

1.) FTDNA kit 438809, the only Yemeni Jewish kit in the J1 y-DNA project (with Levitical status)
2.) MK2015#16887/Israel41 in Karmin 2015 (A recent bottleneck of Y chromosome diversity coincides with a global change in culture)
3.) MK2015#16973/ISR49 in Karmin 2015 (A recent bottleneck of Y chromosome diversity coincides with a global change in culture)

As far as Iranian Kurdistan is concerned one Kurdish Jewish sample from Saqqez speaking Hulaulá language (Jewish Aramaic):

FTDNA #105085 Simani

The only problem with my above assumption is the fact that Mizrachim are heavily undersampled

StillWater
12-25-2019, 02:05 PM
That's an interesting workaround. Hadn't heard of that before, but I could see it!

And most projects I'm interested in at FTDNA have allowed me to join as a guest—except the J-ZS241 one, which expelled me angrily when it found out I was "doing research". I actually haven't tested my Y chromosome at FTDNA, but I transferred my autosomal results from 23andMe to FTDNA's Family Finder several years back.

https://en.wikipedia.org/wiki/Isaak_Asknaziy
https://en.wikipedia.org/wiki/David_Ashkenazi
https://en.wikipedia.org/wiki/Ludv%C3%ADk_A%C5%A1kenazy

Ashkenazi->Sephardi->Ashkenazi? Or a latecomer from Germany to Eastern Europe? I suppose Ashkenazi->Krymchak->Ashkenazi is also possible, but not likely in the cases where they lived far away from Crimea.

hartaisarlag
12-25-2019, 05:40 PM
In my opinion a substantial number of Oriental Jews will turn out FGC3723+ which happens to be my J1-subclade.

As far as Yemen is concerned there are only three Jewish J1 samples (all three FGC3723+):

1.) FTDNA kit 438809, the only Yemeni Jewish kit in the J1 y-DNA project (with Levitical status)
2.) MK2015#16887/Israel41 in Karmin 2015 (A recent bottleneck of Y chromosome diversity coincides with a global change in culture)
3.) MK2015#16973/ISR49 in Karmin 2015 (A recent bottleneck of Y chromosome diversity coincides with a global change in culture)

As far as Iranian Kurdistan is concerned one Kurdish Jewish sample from Saqqez speaking Hulaulá language (Jewish Aramaic):

FTDNA #105085 Simani

The only problem with my above assumption is the fact that Mizrachim are heavily undersampled

That connection is interesting. Wonder how many lines we could expect Yemenites to share with Mesopotamian Jews, as opposed to other Jewish groups. Also, IIRC there are Yemenite kohanim under J-Z18271?

Either way, the undersampling of non-Ashkenazim is really a shame. I’m no more interested in Ashkenazi Y chromosomes than those of any other Jewish group - but there’s no other Jewish group for which anything like this is remotely possible.

Shamash
12-25-2019, 07:43 PM
That connection is interesting. Wonder how many lines we could expect Yemenites to share with Mesopotamian Jews, as opposed to other Jewish groups. Also, IIRC there are Yemenite kohanim under J-Z18271?

As far as I know there aren't, the only Yemenite J-Z18271 sample has no recent Jewish background but is a genetic Cohen, FTDNA kit 476158, Al-Abdali from Wadi Yahar, عبدالله العبدلي العبادل-وادي يهر , J-Z18271, negative for the downstream SNPs: ZS237, S12192 & Z18290. But there might well be untested samples waiting to be uncovered.

sdrucker
12-26-2019, 03:30 AM
For the following branches, there is circumstantial evidence of either a Sephardic or Italian Jewish connection, and if I were a betting man, I’d put money on the connection being genuine. That said, in these cases there is 1) no structural evidence of a split, or muddled evidence of such a split, and 2) no means by which to date a hypothetical split.

#2, J2a-L556 - Possible SB: Hispanic

J2a-L556 (TMRCA: 850 CE) has according to FTDNA been found a Mexican, a Chilean, and a Honduran (as well as 2 Italians and a Pakistani) as a terminal SNP.

#3, G2b-Y12975 - PSB: Italian Jew

G2b-Y12975 (TMRCA: 900 CE) has apparently been found in a single-SNP YSEQ test in an Italian Jew with deep roots in Tuscany, and prior to that, Bologna.

#4, Q1b-Y2200 - PSB: Greek/Hispanic/Italian

Q1b-Y2200 (/Q1b-Y2232) (TMRCA: 450 CE) splits into two main branches; one of them, Q1b-Y2197, has a primary split, likely earlier than 700 CE, between the Ashkenazi Q1b-Y2198, and the Greek/Italian/Portuguese/Brazilian Q1b-BZ72. This latter branch might be of Sephardic, Italki, or Romaniote provenance (and the age of Q1b-Y2200 implies a pre-Ashkenazi presence in the Western Jewish community, unless it’s a late classical non-Jewish introgression), but I’ve been able to find no information about these individuals.

#7, J2a-Y15223 - PSB: Hispanic, Egyptian Jew?

J2a-Y15223 (TMRCA: 550 CE) without downstreams has been confirmed in a Mexican, according to FTDNA, and estimated in an Egyptian Jew.

#13, J1-Y6384 - PSB: Turkish? Puerto Rican?

J1-Y6384 (TMRCA: 450 CE) is split twice vertically at FTDNA, but not at YSEQ. The culprits: a Turkish kit (Sephardic?) and upstream, a Puerto Rican kit.

#15, E1b-Y35934 - PSB: Italian Jew

E1b-Y35934 (TMRCA: 1250 CE) was found by Behar et al. 2017 in an Italian Jew—though without a pedigree, it’s impossible to say what community that Italian Jew belongs to.

#35, J1-ZS2597 - PSB: Hispanic

J1-ZS2597 (TMRCA: 50 CE) has two basal individuals (who apparently don’t cluster with each other)—a Mexican and a Ukrainian (Jew?).


Hello,
I'm J1-ZS2597 (part of AB-30a on JewishDNA Net. ZS2606 on FTDNA) as of the YFull tree published on 4/2019. Not sure if I'm the Jew of Ukrainian origin you're mentioning, but the most recent version of the tree has (along with me) a Mexican-American that can trace their family history to a mestizo named de Udave in the late 17th century, as well as another Galicia Jew and a man named "Leppo". The Udave name may go back to the town of Udave, which was near a major Jewish settlement in Pamplona, the former capital of Navarre, up through the 14th century.

We'll never know for sure, but on FTDNA three of us I mentioned also group with a known descendant of crypto-Jews ("Xuetas") from Mallorca that converted to Judaism and became a rabbi after making aliyah in Israel, on FTDNA's Xueta project. I personally don't have any family lore about my patrilineal line beyond coming from Burztsyn in Galicia in what's now Ukraine, but I believe the Iberian origin given what I've been finding after doing the Big Y test. I also have a number of Hispanic surnamed matches on 67 marker testing on FTDNA at a distance of 4 and 5. Most of the testers at this level seem to have stopped at 67 marker testing, though, so I don't know more about many of them than a basic M267 haplogroup read.

I'm 85%+ Ashkenazi on the usual suspects, with 5% to 10% various mixes of Greek/Italian, Southeast Europe, and the like.

Hope that helps - and if there's any more information I can provide please let me know.

Stuart
YF19121 on YFull, B268154 on FTDNA

StillWater
12-26-2019, 04:21 AM
Hello,
I'm J1-ZS2597 (part of AB-30a on JewishDNA Net. ZS2606 on FTDNA) as of the YFull tree published on 4/2019. Not sure if I'm the Jew of Ukrainian origin you're mentioning, but the most recent version of the tree has (along with me) a Mexican-American that can trace their family history to a mestizo named de Udave in the late 17th century, as well as another Galicia Jew and a man named "Leppo". The Udave name may go back to the town of Udave, which was near a major Jewish settlement in Pamplona, the former capital of Navarre, up through the 14th century.

We'll never know for sure, but on FTDNA three of us I mentioned also group with a known descendant of crypto-Jews ("Xuetas") from Mallorca that converted to Judaism and became a rabbi after making aliyah in Israel, on FTDNA's Xueta project. I personally don't have any family lore about my patrilineal line beyond coming from Burztsyn in Galicia in what's now Ukraine, but I believe the Iberian origin given what I've been finding after doing the Big Y test. I also have a number of Hispanic surnamed matches on 67 marker testing on FTDNA at a distance of 4 and 5. Most of the testers at this level seem to have stopped at 67 marker testing, though, so I don't know more about many of them than a basic M267 haplogroup read.

I'm 85%+ Ashkenazi on the usual suspects, with 5% to 10% various mixes of Greek/Italian, Southeast Europe, and the like.

Hope that helps - and if there's any more information I can provide please let me know.

Stuart
YF19121 on YFull, B268154 on FTDNA

See Kevin Brooks' article on Sephardim in Galicia(Poland and Ukraine). Have you either tested with AncestryDNA or 23andMe? If so, what did they give you for your ethnicity composition?

sdrucker
12-26-2019, 04:29 AM
See Kevin Brooks' article on Sephardim in Galicia(Poland and Ukraine). Have you either tested with AncestryDNA or 23andMe? If so, what did they give you for your ethnicity composition?

I think I saw the Brooks article linked on the Sephardic Poland FB group, but should reread it. I came up with 100% Ashkenazi on Ancestry after their last update. Never did 23andme, but DNA.Land has me at 87% Ashkenazi, 7.7% "Mediterranean Islander", "North Slavic" 3.6%, and the rest Southwestern European for what it's worth.

StillWater
12-26-2019, 04:32 AM
I think I saw the Brooks article linked on the Sephardic Poland FB group, but should reread it. I came up with 100% Ashkenazi on Ancestry after their last update. Never did 23andme, but DNA.Land has me at 87% Ashkenazi, 7.7% "Mediterranean Islander", "North Slavic" 3.6%, and the rest Southwestern European for what it's worth.

DNA.LAND is garbage. What genetic communities did AncestryDNA assign you? Their Jewish category isn't just Ashkenazi, by the way.

sdrucker
12-26-2019, 04:39 AM
DNA.LAND is garbage. What genetic communities did AncestryDNA assign you? Their Jewish category isn't just Ashkenazi, by the way.

Ancestry's latest estimate:
European Jewish - 100%
Western & Central Europe
Western Ukraine, Moldova & Eastern Romania
Central & Eastern Europe
Lithuania, Latvia, Ukraine & Western Russia
Northeast Poland, Lithuania, Latvia & West Belarus

Nothing more specific than this. Maybe saying "100% Ashkenazi" is too much a simplification of these results, though. Other than that I'm Jewish from these areas for all four grandparents, so it's kind of stating the obvious through recent generations.

StillWater
12-26-2019, 11:32 PM
How did you all miss this?

https://khazardnaproject.wordpress.com/2019/09/14/the-origin-of-the-shapiro-rabbinical-lineage-according-to-the-y-dna/


The potential Alan/Khazar of the Shapiro Rabbinical Lineage is discussed in a draft manuscript by Paull, J.M, Briskman, J., and Steeble, S.K.


In other words, based on ancient DNA at least one of the “Four Jewish mtDNA mothers” is from Europe, as predicted by Costa et al. (2013). I agree with the authors that it is unlikely that the remaining haplogroups are from the Middle East. This work was made possible using ancient mtDNA and the construction of ancient mtDNA databases. We are working to map ancient Y chromosomes to create an equivalent resource.


From the study Elhaik quotes:


One possibility may be linked to the Alans – Khazars connections. G2a1a samples have been found in recent paleo DNA studies of the Alans and one of them was from the Saltovo-Mayaki Culture closely associated with the Khazar Khaganate.

StillWater
12-26-2019, 11:55 PM
This is the lineage in the study Elhaik referenced: G-M201 > P15 > FGC7535 > FGC595/L293 > FGC693 > FGC715 > FGC1160 > Z6673 > FGC1107

StillWater
12-27-2019, 12:29 AM
Most members seem to be Litvaks so far and Shapiro is an overwhelmingly Litvak name.

Austria
Lithuania
Belarus
Belarus
Belarus
Russian Federation(probably Belarus)
Belarus
Latvia

Lithuania
Ukraine
Unknown Origin
Lithuania
Lithuania
Ukraine
Belarus

According to this, Lezgins and Karachays seem to be the closest members thus far:
http://forum.molgen.org/index.php?topic=4371.120;wap

Other sources say Ossetians are.

According to this: https://www.familytreedna.com/public/G-L293/default.aspx?section=yresults , the closest seem to be someone from North Ossetia, Balkaria and a Kumyk.

Not that I'm sold on this, but both Ethnogene and 24Genetics gave me trace amounts of Ossetian. Of course, what matters here and what I can't find - the MRCA estimates. What is the MRCA estimate for the branch uniting the Caucasians and Ashkenazim? How old is the Ashkenazi branch(one source says Middle Ages)?

hartaisarlag
12-27-2019, 12:29 AM
This is the lineage in the study Elhaik referenced: G-M201 > P15 > FGC7535 > FGC595/L293 > FGC693 > FGC715 > FGC1160 > Z6673 > FGC1107

Doesn't make JewishDNA.net's Top 50 (again, this is a squishy designation), but interesting. Sure looks North Caucasian. Can't find any TMRCA figures either.

StillWater
12-27-2019, 12:44 AM
Doesn't make JewishDNA.net's Top 50 (again, this is a squishy designation), but interesting. Sure looks North Caucasian. Can't find any TMRCA figures either.

It should be noted that there is a Lebanese for G-Z6653*.

StillWater
12-27-2019, 12:50 AM
It might be this one: AB-060 G-L293-Z6673 4 1074-1695 below G A

http://jewishdna.net/AB-060.html

hartaisarlag
12-27-2019, 12:54 AM
It might be this one: AB-060 G-L293-Z6673 4 1074-1695 below G A

http://jewishdna.net/AB-060.html

Believe it's this one, but not sure what the basis for that calculation is.

hartaisarlag
01-02-2020, 07:52 AM
As for TMRCAs, I am relying on YFull's latest estimates.

The larger the branch, generally, the more trustworthy the TMRCA estimate—for obvious reasons. For many of the smaller branches in JewishDNA.net's top 50, only 2 or 3 individuals have been uploaded to YFull, which yields a good handful of post-1250 TMRCAs, when the actual figure is probably earlier. Setting these aside, with very few exceptions, nearly all the branches have TMRCAs between 350 and 1050 CE. To be clear, there are a good number of branches which do not appear clearly on YFull, so naturally, YFull does not offer TMRCAs for them.

Half of the top 10 have TMRCAs between 750 (this is E-Y6938, and it keeps oscillating between 750 and 800) and 900 CE, which indicates an origin within a few generations of the first documented Ashkenazi communities in the Rhineland and Alsace-Lorraine.

Only 4 of the top 50 branches have earlier TMRCAs:

- #6, J1-Y3088/Z18271, as we know, is a special case with a pre-classical TMRCA, even though its largest Ashkenazi sub-branch (if YFull is to be trusted) has a TMRCA of 400 CE.
- #29, J2b-Y33795 (downstream of J2b-L283), also a Kohen branch, has a TMRCA of 200 CE, which given its likely Italian origin, is mighty suggestive of a classical Roman conversion/rape/NPE.
[Given the TMRCA, I expect that Jewish non-Ashkenazim will show up in this branch. One sample, labeled with a Dutch flag on YFull, claims Sephardic descent, but shares a TMRCA of 1000 CE with an Ashkenazi.]
- #35, J1-ZS2597 (downstream of J1-Y18297) has a TMRCA of 50 CE.
[Given the TMRCA, I expect that Jewish non-Ashkenazim will show up in this branch. The branch contains an ungrouped Mexican individual and an ungrouped Ashkenazi. The branch's nearest sibling, from which it diverges ca. 900 BCE, contains several Palestinians.]
- #50, J1-ZS1686 (downstream of J1-YSC76) has a TMRCA of 250 CE.
[Given the TMRCA, I expect that Jewish non-Ashkenazim will show up in this branch. The branch also diverges from a single Mexican individual ca. 250 CE.]

I either missed one of these, or the latest update to YFull has newly cast it in relief.

#39, J2a-FGC4975, seems to have been the subject of a recent upload drive by the Avotaynu project, and apparently has a TMRCA of 0. I believe JewishDNA.net incorrectly labels only one of its two sub-branches, J2a-L254, as an Ashkenazi branch.

This one is a decent candidate for an Ashkenazi J2a branch with a European Mediterranean origin, though it's hard to say definitively.

StillWater
01-03-2020, 11:24 PM
This is the lineage in the study Elhaik referenced: G-M201 > P15 > FGC7535 > FGC595/L293 > FGC693 > FGC715 > FGC1160 > Z6673 > FGC1107

Hoping to hear Agamemnon's and Principe's opinion on this.

Claudio
01-05-2020, 07:25 PM
I either missed one of these, or the latest update to YFull has newly cast it in relief.

#39, J2a-FGC4975, seems to have been the subject of a recent upload drive by the Avotaynu project, and apparently has a TMRCA of 0. I believe JewishDNA.net incorrectly labels only one of its two sub-branches, J2a-L254, as an Ashkenazi branch.

This one is a decent candidate for an Ashkenazi J2a branch with a European Mediterranean origin, though it's hard to say definitively.

Hey Hartaisarlag,any chance you could post a similar thread,but on your present current observations regarding the origins of the various MtDNA of Ashkenazim?

hartaisarlag
01-05-2020, 08:50 PM
Hey Hartaisarlag,any chance you could post a similar thread,but on your present current observations regarding the origins of the various MtDNA of Ashkenazim?

I would, but I don’t know anywhere near as much about them. Great thought, though. Any volunteers?

StillWater
01-05-2020, 09:09 PM
I would, but I don’t know anywhere near as much about them. Great thought, though. Any volunteers?

I've made quite a few posts regarding MTDNA in the other thread. I even found a source suggesting that J1c7a entered eastern Euros through Jews. I should find it again.

Principe
01-05-2020, 10:53 PM
Hoping to hear Agamemnon's and Principe's opinion on this.

Honestly its a small lineage, something that Hartaisarlag pretty much said, likely Northern Caucasus>Eastern Ashkenazim introgression, could also have been a convert from the Byzantine Empire, won’t likely be a lineage amongst the founding Ashkenazim gene pool.

StillWater
01-05-2020, 11:23 PM
Honestly its a small lineage, something that Hartaisarlag pretty much said, likely Northern Caucasus>Eastern Ashkenazim introgression, could also have been a convert from the Byzantine Empire, won’t likely be a lineage amongst the founding Ashkenazim gene pool.

The only kinda introgression I can imagine is if there was a soldier in the Russian Empire of North Caucasian descent and then rape followed. North Caucasians didn't live in the Pale or in the Polish-Lithuanian commonwealth, from what I know. I just recalled that Kevin Brook once mentioned it before: someone with in this haplogroup says that the FTDNA calculator pointed to Ashkenazim and North Caucasians sharing a common ancestor in the 11th century. The Byzantine convert is possible. Can't find where the Ashkenazim should be: https://yfull.com/tree/G-Z6653/ . The basal Lebanese does make wonder if it could be Judean.

hartaisarlag
01-05-2020, 11:36 PM
I've made quite a few posts regarding MTDNA in the other thread. I even found a source suggesting that J1c7a entered eastern Euros through Jews. I should find it again.

Where is that? Can't imagine how it would be reconciled with presence in Iron Age Poland (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5802798/). I know it's been found in Iranians, but until we have better info on phylogenetic structure and branch age, I'm not sure what to make of that datum.

hartaisarlag
01-05-2020, 11:48 PM
The basal Lebanese does make wonder if it could be Judean.

All of G-FGC1160, with the exception of a Portuguese or two, seems to be Caucasian (and mostly North Caucasian). Ashkenazim in the branch are nested within this variation. Unless these Caucasians too are of Judean origin, I highly doubt it. Basal Lebanese splitting from the rest of the branch ca. 4800 CE could be tied to anything.

hartaisarlag
01-05-2020, 11:49 PM
Honestly its a small lineage, something that Hartaisarlag pretty much said, likely Northern Caucasus>Eastern Ashkenazim introgression, could also have been a convert from the Byzantine Empire, won’t likely be a lineage amongst the founding Ashkenazim gene pool.

And hey, could actually be a Khazar convert. Genetic data don't eliminate the possibility that this happened a few times, and had an autosomally undetectable, but uniparentally traceable, impact.

Agamemnon
01-06-2020, 12:05 AM
Doesn't make JewishDNA.net's Top 50 (again, this is a squishy designation), but interesting. Sure looks North Caucasian. Can't find any TMRCA figures either.

More likely to be South Caucasian/Kartvelian than North Caucasian though.

Not sure you've noticed, but Z18271's (Y3088) TMRCA estimates are back to ~2,900 yBP. You can expect to keep going down with more results, same thing with S17446 which is ~2,100 years old now.

hartaisarlag
01-06-2020, 12:07 AM
More likely to be South Caucasian/Kartvelian than North Caucasian though.

Not sure you've noticed, but Z18271's (Y3088) TMRCA estimates are back to ~2,900 yBP. You can expect to keep going down with more results, same thing with S17446 which is ~2,100 years old now.

I did notice that! IIRC S17446 came down 1k years, which makes much more sense. By the way, who are the two new J-Y3088* samples (one labeled Syrian, one unlabeled)?

Also, what gives you the sense of Kartvelian specifically? I know very little about uniparentals in Georgia and the North Caucasus, or about sampling biases in NGS Y-testing in those regions.

hartaisarlag
01-06-2020, 12:22 AM
Another interesting note: the TMRCA of the famous (G-M377 >) G-Y12975 has been moved up to ca. 950 CE. This is now the only large Ashkenazi founding branch whose TMRCA perfectly matches the founding date of the proto-Ashkenazi community. It also happens that we haven't found any secure Sephardic connection for it, and that an old north/central Italian Jewish family seems to carry it too. If they happened to split the branch at the root, that would give the cleanest migration map/timeline yet for any of these branches.

Now, I could be wrong in trusting the precision of these estimates too much, but in the case of many of these branches, they seem not to have moved much even with the addition of lots of internal structure over the years, especially relative to one another. But my impression is that 950 is substantively different from, say, 750 in the case of E-Y6938. (As are the major branches with a TMRCA of 450-500, more obviously.)

A TMRCA of 950 at least theoretically allows us to pinpoint the origin of the line in the early Ashkenazi community in one particular location immediately prior to migration to Germany; all of its internal structure would postdate the migration to early Ashkenaz. On the other hand, a TMRCA of 750, with certain sub-branches dating as far back as 750, implies a more complex history, and strongly suggests that certain sub-branches began their spread separately before the migration to Ashkenaz—and that the migration to Ashkenaz was likely undertaken by multiple individuals in multiple sub-branches of E-Y6938, who one assumes all lived in the same regional Jewish community between ca. 750 and ca. 950.

Agamemnon
01-06-2020, 12:51 AM
I did notice that! IIRC S17446 came down 1k years, which makes much more sense. By the way, who are the two new J-Y3088* samples (one labeled Syrian, one unlabeled)?

Also, what gives you the sense of Kartvelian specifically? I know very little about uniparentals in Georgia and the North Caucasus, or about sampling biases in NGS Y-testing in those regions.

I have no idea who the unlabeled individual is, however the Syrian sample is almost undoubtedly from the Anzaroot family (also present in Lebanon). Either that or from the Tawil family (doubtful), both are of priestly descent.

There are two things that suggest to me that this is Kartvelian. First off its distribution is pan-Kartvelian, it is found in Mingrelian, Laz and Svan speakers, in comparison it encompasses a wide array of ethnic groups in the North Caucasus (Kabardian, chiefly Ossetian, Ingush, etc). Second, going off the ancient DNA record so far, it's clear that Z6653's branches spread from the South Caucasus during the Bronze Age.

hartaisarlag
01-06-2020, 12:59 AM
There might really be something to it, if we look genealogically and at trustworthy family trees we can confidently list 20 Ashkenazi Y lines originating from the Rhineland-Palatine communities and these 20 lines combined easily make over 50% (probably at the 52-55% range).

J2a-L556 (2nd largest Y line)
G-M377 (3rd)
Q-Y2200 (4th)
J1-Y5400 (6th)
E-M84>Y14891 (7th)
J2a-Y15223 (8th)
J1-L816 (9th)
J2a-Z6048>Y36257 (11th)
J1-ZS8615 (12th)
E-PF1952 (14th)
E-PF1974 (16th)
E-V22>BY7500 (19th)
R1b-V88>FGC20980 (20’s range)
R1b-U152>L2>L408 (20’s range)
R1b-Z2103>Y19847 (20’s range)
J2b-L283>CTS6190 (30’s range)
G-M406>S11415
R1b-L584>FGC14600
R1b-L21>Z18106
J1-PF7263

The last four are still in the top 50

OK, another thing I've been thinking about, after discussing it a number of times with my friend who focused on E-M84 > Y14891: my line, E-Y6938, is not attested in old Rhineland families, and it has virtually no recent Rhineland roots in its known ranks—compared to several of the other top few lineages. A simple t-test comparing the two large Ashkenazi E-M34 branches' % Western Ashkenazi basically says there's a 98% chance that the difference is not due to random differences in sampling.

Does this mean that E-Y6938 bypassed the Rhineland and entered the early Central/Eastern European Jewish community via some other route? It's tempting to start lining up evidence for these distinctions.

But here's the thing: what's interesting about E-Y6938 is that its pre-Ashkenazi origins ca. late classical times are among the most clearly constrained of the large Ashkenazi branches. Circa 400 CE, it splits from a non-Ashkenazi branch—whether or not to call it Sephardic is an open question, but I think it could be safely called "West/Central Mediterranean Jewish". This makes an origin via Mizrahi or Byzantine Jews in Bohemia or some point farther east virtually impossible—the Jewish migrational history of this branch is firmly anchored in the west (and to anyone commenting on the Turkish flag in YFull, this is a Tunisian Jew with claimed Turkish Sephardic origins; I'm not willing to fully credit these claims).

The only easy-to-imagine migration route that accommodates these two facts—seeming absence from Rhenish and Alsatian Jews and deep roots in the western half of the empire—is one from Italy to Regensburg/Vienna/Prague. Would it ever be possible to prove this? I'm doubtful. But spending enough time with the patterns has me feeling bold.

EDIT: no evidence for Jews in Vienna prior to the 12th century. But, Jews are attested in Magdeburg and Merseburg in eastern Germany in the late 10th century, and in the 11th century in Meissen (Upper Saxony) and Bamberg (Upper Franconia). There's also arguable evidence for Jewish-specific tolls in/around Passau, in southeastern Bavaria, in the early 10th century, but it could be chalked up to a few itinerant merchants, rather than any nearby permanent community.

Claudio
01-06-2020, 01:02 AM
The only kinda introgression I can imagine is if there was a soldier in the Russian Empire of North Caucasian descent and then rape followed. North Caucasians didn't live in the Pale or in the Polish-Lithuanian commonwealth, from what I know. I just recalled that Kevin Brook once mentioned it before: someone with in this haplogroup says that the FTDNA calculator pointed to Ashkenazim and North Caucasians sharing a common ancestor in the 11th century. The Byzantine convert is possible. Can't find where the Ashkenazim should be: https://yfull.com/tree/G-Z6653/ . The basal Lebanese does make wonder if it could be Judean.

Not necessarily always Rape:
35745

Principe
01-06-2020, 01:41 AM
OK, another thing I've been thinking about, after discussing it a number of times with my friend who focused on E-M84 > Y14891: my line, E-Y6938, is not attested in old Rhineland families, and it has virtually no recent Rhineland roots in its known ranks—compared to several of the other top few lineages. A simple t-test comparing the two large Ashkenazi E-M34 branches' % Western Ashkenazi basically says there's a 98% chance that the difference is not due to random differences in sampling.

Does this mean that E-Y6938 bypassed the Rhineland and entered the early Central/Eastern European Jewish community via some other route? It's tempting to start lining up evidence for these distinctions.

But here's the thing: what's interesting about E-Y6938 is that its pre-Ashkenazi origins ca. late classical times are among the most clearly constrained of the large Ashkenazi branches. Circa 400 CE, it splits from a non-Ashkenazi branch—whether or not to call it Sephardic is an open question, but I think it could be safely called "West/Central Mediterranean Jewish". This makes an origin via Mizrahi or Byzantine Jews in Bohemia or some point farther east virtually impossible—the Jewish migrational history of this branch is firmly anchored in the west (and to anyone commenting on the Turkish flag in YFull, this is a Tunisian Jew with claimed Turkish Sephardic origins; I'm not willing to fully credit these claims).

The only easy-to-imagine migration route that accommodates these two facts—seeming absence from Rhenish and Alsatian Jews and deep roots in the western half of the empire—is one from Italy to Regensburg/Vienna/Prague. Would it ever be possible to prove this? I'm doubtful. But spending enough time with the patterns has me feeling bold.

EDIT: no evidence for Jews in Vienna prior to the 12th century. But, Jews are attested in Magdeburg and Merseburg in eastern Germany in the late 10th century, and in the 11th century in Bamberg. There's also arguable evidence for Jewish-specific tolls in/around Passau, in southeastern Bavaria, in the early 10th century, but it could be chalked up to a few itinerant merchants, rather than any nearby permanent community.

Initially when I wrote that post I always had the inclination to believe E-Y6938 and R-Y2619 entered from a different route, I was think along the lines that both those became strong in the early Bohemian communities.

Serious question could it be possible that there were several distinct communities that contributed to the Ashkenazim ethnogensis, let’s say North-Central Italy and Provençal/South France being locks, could the Balkans/Byzantine Empire also serve as one of the founding communities? Maybe there is direct evidence with Romaniote>Ashkenazi with J2a-P244 lineage?

StillWater
01-06-2020, 01:46 AM
Not necessarily always Rape:
35745

Good find, but Pale Jews were ultra-orthodox and Yiddish-speaking at the time. Mind you, this clade is famous for being found in a Rabbincal family. The possibility here implies that a Jewish woman pretended to do something to her family, for which she was permitted to enter Russia for, while in fact being a prostitute. She then would've had to have been matched off for marriage after a "work encounter", while not being visibly pregnant. The other option is that her parents raised the kid and eventually her progeny was able to marry other Jews. However, given the dates, it should be even more rare than it is now and traceable. As of right now, it's found among Jews from different countries, with completely unrelated surnames. As far as such a date is concerned, this is a more likely venue:


With the third partition of Poland (1795), the law was also applied to the provinces of *Vilna and *Grodno. In 1799 *Courland was added to the Pale of Settlement. In the "Jewish Statute" promulgated in 1804, the province of Astrakhan and the whole of the northern Caucasus were added to the regions open to Jews. In 1812, upon its annexation, *Bessarabia was also included. The "Kingdom of Poland," incorporated into Russia in 1815, which included ten provinces that later became known as the "Vistula Region," was not officially included within the Pale of Settlement, and until 1868 the transit of Jews through it to the Lithuanian and Ukrainian provinces was prohibited by law. In practice, however, the provinces of the Vistula Region were generally included within the Pale of Settlement.

To sum up, it was the intention of the Russian legislators of the reigns of Catherine II and Alexander I to extend the Pale of Settlement beyond the regions acquired from Poland only to those areas where Jews could serve as a colonizing element. However, from the reign of Alexander II the restrictive aspects of the Pale of Settlement became accentuated, for while freedom of movement for non-Jews in Russia increased, in particular after the emancipation of the serfs, the restrictions on the movement of Jews beyond the Pale remained in force, and became explicitly underlined within the Pale itself. This was accomplished both by anti-Jewish enactments on the part of the government and by the growing impatience of Jewish society and liberal public opinion with these disabilities.

Czar Nicholas I (under whom the term "Pale of Settlement" was coined) removed Courland from the Pale in 1829; however, the rights of the Jews already settled and registered there were maintained. In 1835 the provinces of Astrakhan and the northern Caucasus were excluded from the Pale. In 1843 Nicholas I ordered the expulsion of the Jews from a strip of 50 versts (about 33 mi.) in width extending along the border with Prussia and Austria. Many difficulties were encountered in the application of this law, and in 1858 it was redrafted to apply only to those Jews who would wish to settle in the border zone after that year. A similar law which had applied to the provinces of Russian Poland (where the border zone closed to Jewish residence was 21 versts in width) was abrogated in 1862. In 1827 severe restrictions were imposed on the residence of Jews in Kiev, the largest town in southern Russia, that served as an important commercial center for the surrounding regions which had a dense Jewish population.

This allows more descendants than an event from the late 1800's, but still likely isn't sufficient to explain the surname/geographical diversity (as narrow as it is) in this clade. It's currently found in what - around 1 in 400 or so Ashkenazim? Sample bias and insufficient total sampling may explain some inflation, but again, the surname and geographical diversity are hard to explain. Also, this would place the North Caucasian ancestor in the autosomally traceable range. Yet, are there Ashkenazim with such cousin matches? Do Ashkenazim score the relevant categories on the big DNA tests? To be fair, I did get trace North Caucasian on both Ethnogene and 24Genetics, but what about something like 23andMe? I've went through pretty much all of mine (checking major ethnic groups, as opposed to individuals on MH) , and maybe had 1 very distant match with someone who was partly of Caucasian origin (I'll have to double check).

I'm wondering if it's through an introgression into Mountain Jews and then a migration to Ashkenazi communities. Would be interesting to see if it shows up in Krymchaks and what the MRCA date is.

StillWater
01-06-2020, 01:54 AM
There are two things that suggest to me that this is Kartvelian. First off its distribution is pan-Kartvelian, it is found in Mingrelian, Laz and Svan speakers, in comparison it encompasses a wide array of ethnic groups in the North Caucasus (Kabardian, chiefly Ossetian, Ingush, etc). Second, going off the ancient DNA record so far, it's clear that Z6653's branches spread from the South Caucasus during the Bronze Age.

Could this be a rare South Caucasus -> Israel branch, such that we've only found Jews with it, and even there, only a few?

hartaisarlag
01-06-2020, 04:20 AM
Initially when I wrote that post I always had the inclination to believe E-Y6938 and R-Y2619 entered from a different route, I was think along the lines that both those became strong in the early Bohemian communities.

Serious question could it be possible that there were several distinct communities that contributed to the Ashkenazim ethnogensis, let’s say North-Central Italy and Provençal/South France being locks, could the Balkans/Byzantine Empire also serve as one of the founding communities? Maybe there is direct evidence with Romaniote>Ashkenazi with J2a-P244 lineage?

I think it's likely they played some direct role—where else could the pre-Ashkenazi Jews of Eastern Europe have come from? Weinreich seemed convinced of it, but admittedly, didn't supply any clear proof. Plus, both R1b-A11720 and I2-Y23115 point to—if not conclusively—conversions in that realm. There could be J2a and G lines too, but those blend in much more easily.

How do you figure J2a-Z39057 / J2a-P244 indicates it? Maybe I'm just not familiar enough with the branch. But neither the FTDNA Haplotree nor the Romaniote FTDNA project supports JewishDNA.net's claim that the branch is found among Romaniotes. I'll be excited if it is!

As for your original post from back then, I notice that a lot of it corresponds to what I've seen in the Jews of Frankfurt project. Did you use other sources (not doubting the original collection, I've just never gone too deep on this)? Notable to me that R1b-FGC8564/Ivanhoe is also missing, as is the aforementioned R1b-A11720, despite the clear attestation of many smaller branches. It's very possible that Bohemia was an original early seat of E-Y6938 lines, but again, its pre-Bohemian departure point could not have been to the southeast.

Also, synthesizing with my earlier point about 750 vs. 950, it's unusual that none of the several Y6938 > Z36123 lines would have made it to the Rhineland. They had been diverging for up to 3 centuries before settlement in Germany/Austria/Bohemia. There are German Jews attested in the large Y15936 and Y15561 branches, and one is connected to a Rhenish placename—but that line falls under Y15938, which has a TMRCA of ca. 1525, and contains Jews from all over central and eastern Europe—could easily be the result of east > west migration. The one exception to all of this is one of our 3 Z36123* lines (I am one of the others, and the other is AWOL and not a member of our project), whose last known family location is in northern Hesse. This to me raises the possibility of an early migration to Germany, but not to one of the Rhineland centers.

hartaisarlag
01-06-2020, 09:12 AM
Initially when I wrote that post I always had the inclination to believe E-Y6938 and R-Y2619 entered from a different route, I was think along the lines that both those became strong in the early Bohemian communities.

Serious question could it be possible that there were several distinct communities that contributed to the Ashkenazim ethnogensis, let’s say North-Central Italy and Provençal/South France being locks, could the Balkans/Byzantine Empire also serve as one of the founding communities? Maybe there is direct evidence with Romaniote>Ashkenazi with J2a-P244 lineage?

Also, revisiting your list, it appears that 13 of the 20 largest branches are present, with the following absent. The larger the branch, the less likely this is by chance:

R1a-Y2619 < Bavaria/Bohemia/Elbe?
E-Y6923 < Bavaria/Bohemia/Elbe?
R1b-FGC8564 < later arrival from Spain?
R1b-A11720 < southeastern conversion
J2a-FGC21085 < ?
G-FGC228 < ?
J1-FGC5215 < later arrival from Spain?

Principe
01-06-2020, 06:26 PM
I think it's likely they played some direct role—where else could the pre-Ashkenazi Jews of Eastern Europe have come from? Weinreich seemed convinced of it, but admittedly, didn't supply any clear proof. Plus, both R1b-A11720 and I2-Y23115 point to—if not conclusively—conversions in that realm. There could be J2a and G lines too, but those blend in much more easily.

How do you figure J2a-Z39057 / J2a-P244 indicates it? Maybe I'm just not familiar enough with the branch. But neither the FTDNA Haplotree nor the Romaniote FTDNA project supports JewishDNA.net's claim that the branch is found among Romaniotes. I'll be excited if it is!

As for your original post from back then, I notice that a lot of it corresponds to what I've seen in the Jews of Frankfurt project. Did you use other sources (not doubting the original collection, I've just never gone too deep on this)? Notable to me that R1b-FGC8564/Ivanhoe is also missing, as is the aforementioned R1b-A11720, despite the clear attestation of many smaller branches. It's very possible that Bohemia was an original early seat of E-Y6938 lines, but again, its pre-Bohemian departure point could not have been to the southeast.

Also, synthesizing with my earlier point about 750 vs. 950, it's unusual that none of the several Y6938 > Z36123 lines would have made it to the Rhineland. They had been diverging for up to 3 centuries before settlement in Germany/Austria/Bohemia. There are German Jews attested in the large Y15936 and Y15561 branches, and one is connected to a Rhenish placename—but that line falls under Y15938, which has a TMRCA of ca. 1525, and contains Jews from all over central and eastern Europe—could easily be the result of east > west migration. The one exception to all of this is one of our 3 Z36123* lines (I am one of the others, and the other is AWOL and not a member of our project), whose last known family location is in northern Hesse. This to me raises the possibility of an early migration to Germany, but not to one of the Rhineland centers.

When it comes to R1b-A11720 I think a Near Eastern origin for this branch still very highly likely, to date no ancient PF7562 was found, this could be a prime candidate for a Philistine branch, although since we only got 10 of the 108 samples its highly unlikely to know.

As for the J2a-Z39057 there is one member listed on the Romaniote project and the same individual is listed on the Iberian Ashkenaz project.

https://www.familytreedna.com/public/IberianSurnamesofAshkenaz?iframe=yresults

Clearly belonging to the same branch as the Ashkenazim, but we can’t know truly if it was originally Romaniote or Sephardic>Romaniote branch, though we can safely say it comes from another Western Jewish community.

Yes when it came to that post my source was the Jews of Frankfrut site, imo it is one of if not the best genealogical genetic based pedigrees available, and I actually wish others would use the same template. What’s great about this site is that genealogy also helps to place the branches found, so you have some family trees that place ancestors in 1300-1400s coming from other German towns/cities so its traceable to see which of the lineages were there pre Western expansion. Although one issue arrives in general because of the Rhineland Massacres of 1096, its possible that other branches of the top 20 were located there but only the descendants of Eastern Ashkenaz show up, what I’m trying to say its possible that some branches that appear to have TMRCA’s of less than 1000 ybp could have been from the Rhineland communities but because of the massacres only 1 survived and caused such incredibly young TMRCA’s. I’m going to continue with this point with the response to your next post because I think its relevant also with the 950 vs 750 dating.

Principe
01-06-2020, 07:10 PM
Also, revisiting your list, it appears that 13 of the 20 largest branches are present, with the following absent. The larger the branch, the less likely this is by chance:

R1a-Y2619 < Bavaria/Bohemia/Elbe?
E-Y6923 < Bavaria/Bohemia/Elbe?
R1b-FGC8564 < later arrival from Spain?
R1b-A11720 < southeastern conversion
J2a-FGC21085 < ?
G-FGC228 < ?
J1-FGC5215 < later arrival from Spain?

To continue the point further the Rhineland massacres would have potentially created scenarios where some branches could have had older TMRCA’s but these branches were killed off and that’s why younger TMRCA’s exist.

But you see here’s where it gets interesting all the sources of R-Y2619 seem to be coming from the East via Bavaria and Prague, it also seems the Levite family of Epstein seems to have been R1b-L2>L208, so a Bavaria/Bohemia/Elbe origin for Y2619 branch in terms of Community makes sense. In the case of E-Y6938 there are no branches in this community and I think the diversity of subranching would suggest a Rhineland origin is less likely so similarly a Bavaria/Bohemia/Elbe origin is likely for E-Y6938.

I think in the case of R-A11720 and J2a-FGC21085 they follow a similar pattern to Y6938 and Y2619 where they diversified old enough to be also from the Bavaria/Bohemia/Elbe communities as well.

G-FGC228 might be a prime example of how the Rhineland massacres impacted the TMRCA’s or lineage appearance, so I see it possible that FGC228 could have been part of the original Rhineland communities or it could have have come from the Eastern Trio, in all likelihood this branch pre Ashkenazi should have came from either North-Central Italian or Provençal/Southern France Jewish communities.

As for R-FGC8564 and J1-FGC5215, in all honesty yes a later Sephardic arrival makes sense, I know there is no explicit Sephardic origin for the Ivanhoe clade but the Spanish lines appear pretty close to the TMRCA of the Ashkenazim line, so at least an Eastern Germanic origin pre Jewish conversion is almost a given, so if not Spain, Southern France or Italy are likely as well, likely either Visigoth or Suebi in origin, Vandals are likely eliminated because they were already out by 500. The TMRCA of FGC8564 fits perfectly with either or.

Principe
01-06-2020, 07:24 PM
I also think an interesting topic could be you split the top 50 into where they got their start such as Rhineland, Eastern Trio, Sephardic, Mizarchi, Romaniote, not sure yet. Would be a great way to track the paternal ancestors routes!! Truly like a breakdown of Ashkenazi Historical Y dna.

The hardest thing would be to try and differentiate the pre Ashkenazi sources such as North-Central Italy vs Provencal-Southern French communities.

hartaisarlag
01-06-2020, 08:05 PM
When it comes to R1b-A11720 I think a Near Eastern origin for this branch still very highly likely, to date no ancient PF7562 was found, this could be a prime candidate for a Philistine branch, although since we only got 10 of the 108 samples its highly unlikely to know.

As for the J2a-Z39057 there is one member listed on the Romaniote project and the same individual is listed on the Iberian Ashkenaz project.

https://www.familytreedna.com/public/IberianSurnamesofAshkenaz?iframe=yresults

Clearly belonging to the same branch as the Ashkenazim, but we can’t know truly if it was originally Romaniote or Sephardic>Romaniote branch, though we can safely say it comes from another Western Jewish community.

Yes when it came to that post my source was the Jews of Frankfrut site, imo it is one of if not the best genealogical genetic based pedigrees available, and I actually wish others would use the same template. What’s great about this site is that genealogy also helps to place the branches found, so you have some family trees that place ancestors in 1300-1400s coming from other German towns/cities so its traceable to see which of the lineages were there pre Western expansion. Although one issue arrives in general because of the Rhineland Massacres of 1096, its possible that other branches of the top 20 were located there but only the descendants of Eastern Ashkenaz show up, what I’m trying to say its possible that some branches that appear to have TMRCA’s of less than 1000 ybp could have been from the Rhineland communities but because of the massacres only 1 survived and caused such incredibly young TMRCA’s. I’m going to continue with this point with the response to your next post because I think its relevant also with the 950 vs 750 dating.

I'll revise my statement about an obvious SE European introgression story for R1b-A11720, but "very highly likely" Near Eastern seems overconfident in the other direction. I see how the dynamic under cousin branch R1b-FGC42003 seems to imply an SE Europe > Levant introgression, but for now, the branch is nested within the SE European(/ Italian) variation of R1b-Y31335, and has no Near Eastern relatives within the last 4,600 years. There's room for a number of scenarios within that window, but nothing I've seen that clearly points to any Philistine involvement. If we see it in other Jewish groups, I'll revise my weightings.

I now see who the Romaniote in question is, but he doesn't appear to be SNP-confirmed for the branch. That's how I missed it. How certain is it that he's J2a-Z39057?

And gotcha, re: Frankfurt. Your point about the massacres is interesting, and the earlier TMRCAs of R1a-Y2619, E-Y6938, R1b-FGC8564, R1b-A11720, and J2a-FGC21085 definitely jibe with the point you're making.

hartaisarlag
01-06-2020, 08:19 PM
To continue the point further the Rhineland massacres would have potentially created scenarios where some branches could have had older TMRCA’s but these branches were killed off and that’s why younger TMRCA’s exist.

But you see here’s where it gets interesting all the sources of R-Y2619 seem to be coming from the East via Bavaria and Prague, it also seems the Levite family of Epstein seems to have been R1b-L2>L208, so a Bavaria/Bohemia/Elbe origin for Y2619 branch in terms of Community makes sense. In the case of E-Y6938 there are no branches in this community and I think the diversity of subranching would suggest a Rhineland origin is less likely so similarly a Bavaria/Bohemia/Elbe origin is likely for E-Y6938.

I think in the case of R-A11720 and J2a-FGC21085 they follow a similar pattern to Y6938 and Y2619 where they diversified old enough to be also from the Bavaria/Bohemia/Elbe communities as well.

G-FGC228 might be a prime example of how the Rhineland massacres impacted the TMRCA’s or lineage appearance, so I see it possible that FGC228 could have been part of the original Rhineland communities or it could have have come from the Eastern Trio, in all likelihood this branch pre Ashkenazi should have came from either North-Central Italian or Provençal/Southern France Jewish communities.

As for R-FGC8564 and J1-FGC5215, in all honesty yes a later Sephardic arrival makes sense, I know there is no explicit Sephardic origin for the Ivanhoe clade but the Spanish lines appear pretty close to the TMRCA of the Ashkenazim line, so at least an Eastern Germanic origin pre Jewish conversion is almost a given, so if not Spain, Southern France or Italy are likely as well, likely either Visigoth or Suebi in origin, Vandals are likely eliminated because they were already out by 500. The TMRCA of FGC8564 fits perfectly with either or.

I agree with everything here. Weird that an Italian/French/German convert line would gain Levite status, but then again, we see it in the earlier J2b-Y33795 conversion and Kohen status.

G-FGC228 is a bit of an enigma farther down the line (were it not for the Lebanese member upstream from it, we'd have reason to believe it was a North African introgression, maybe via Spain), but any of the scenarios you describe are possible.

I have generally not been inclined to believe that R-FGC8564 reflects a late arrival in the Ashkenazi community, given its TMRCA and the absence of Sephardim in the branch, but the absence from Rhineland pedigrees raises the possibility. Again, were it not for the immediate Iberian cousins, there'd be no reason to even factor Iberia into the equation. J1-FGC5215 already smelled of it pretty strongly, though.

Re: TMRCA's, though, if 500-750 is closer to an "authentic" pre-massacre TMRCA for most of these lines, it would seem unreasonable to expect them to take on such distinctive patterns of distribution (again, most actually fall into the Rhineland paradigm neatly)—any talk of continuous Jewish communities in Germany prior to the 10th century is purely speculative, so by 950 or 1000, we'd be talking multiple distantly-related arrivals for each branch. For R1a-Y2619 and E-Y6938 to buck this trend would seem to imply that these branches, with a few hundred years of structure under their belt, had begun diversifying somewhere other than where the other large branches did—otherwise, you'd expect at least 1 or 2 sub-branches to get swept up in the whole Rhineland thing. But again, we're working with limited data, and it's very possible that there were E-Y6938 lines in, say, Mainz, but just many fewer than one would back-predict from the branch's current frequency.

hartaisarlag
01-06-2020, 08:24 PM
I also think an interesting topic could be you split the top 50 into where they got their start such as Rhineland, Eastern Trio, Sephardic, Mizarchi, Romaniote, not sure yet. Would be a great way to track the paternal ancestors routes!! Truly like a breakdown of Ashkenazi Historical Y dna.

The hardest thing would be to try and differentiate the pre Ashkenazi sources such as North-Central Italy vs Provencal-Southern French communities.

I think this conversation is evolving into that. We'll get there.

I actually wanted to tackle the Italy vs. Provence thing first, and I think there are some hints for some lines (for example, divergences ca. 800-900 from Sephardic branches), but there's really no remnant population with 1,200-year-old pedigrees that would allow us to verify this. The Frankfurt project really is a treasure, in that sense. I would love to see as many Italkim as possible tested; if the G-Y12975 Italki line (with a pedigree going back to 14th century Bologna, IIRC) turns out to be basal, then we really have something to say about that branch. J-L816 > J-ZS2728* among Apulian-origin Romaniotes, diverging ca. 900 CE from a large Ashkenazi branch, makes regional Italian speculation possible too in that case.

Principe
01-06-2020, 08:37 PM
I'll revise my statement about an obvious SE European introgression story for R1b-A11720, but "very highly likely" Near Eastern seems overconfident in the other direction. I see how the dynamic under cousin branch R1b-FGC42003 seems to imply an SE Europe > Levant introgression, but for now, the branch is nested within the SE European(/ Italian) variation of R1b-Y31335, and has no Near Eastern relatives within the last 4,600 years. There's room for a number of scenarios within that window, but nothing I've seen that clearly points to any Philistine involvement. If we see it in other Jewish groups, I'll revise my weightings.

I now see who the Romaniote in question is, but he doesn't appear to be SNP-confirmed for the branch. That's how I missed it. How certain is it that he's J2a-Z39057?

And gotcha, re: Frankfurt. Your point about the massacres is interesting, and the earlier TMRCAs of R1a-Y2619, E-Y6938, R1b-FGC8564, R1b-A11720, and J2a-FGC21085 definitely jibe with the point you're making.

Your totally right my claim does seem over confident, a Eastern Balkan origin for its upstream seems likely, so perhaps a Phyrgian or maybe even a Western Anatolian origin is possible although Thracian origin could be the ultimate origin, issue with modern samples is the huge bias towards Europe. Nevertheless I think we agree A11720 is a fascinating lineage.

Your absolutely right, but via Str’s he is very close to the Z39057 branch, unless he belongs to another L70 branch its possible.

Yeah I think so too :P I would definitely like to see other branches of the top 50 to see if they also match this pattern.

hartaisarlag
01-06-2020, 08:55 PM
Also, revisiting your list, it appears that 13 of the 20 largest branches are present, with the following absent. The larger the branch, the less likely this is by chance:

R1a-Y2619 < Bavaria/Bohemia/Elbe?
E-Y6923 < Bavaria/Bohemia/Elbe?
R1b-FGC8564 < later arrival from Spain?
R1b-A11720 < southeastern conversion
J2a-FGC21085 < ?
G-FGC228 < ?
J1-FGC5215 < later arrival from Spain?

Or, another way of looking at it: of these, 3 of 7 are constrained by West Mediterranean ties within ~500 years of the emergence of Ashkenaz, while 4 aren't.

The three "western" ones are:

E-Y6938 (North African Jewish/Sephardic/Iberian sister branch, split ca. 400 CE)
R1b-FGC8564 (Iberian sister-branch, split ca. 500 CE; definitely a convert line of ultimately Germanic origin)
J1-FGC5215 (Two Iberian sister-branches [almost certainly converso], split ca. 350 CE)

I would rule out a Romaniote or Mizrahi > K'naan origin for these branches. Non-Rhineland but still Spanish/French/Italian in origin is actually pretty good.

As for:

R1a-Y2619
R1b-A11720
J2a-FGC21085
G-FGC228

The possibility is entirely open (though nothing explicitly supports it, over say, an Italy > Regensburg/Prague route). Let's keep an eye out for Romaniote or Mizrahi connections.

Principe
01-07-2020, 01:18 AM
Or, another way of looking at it: of these, 3 of 7 are constrained by West Mediterranean ties within ~500 years of the emergence of Ashkenaz, while 4 aren't.

The three "western" ones are:

E-Y6938 (North African Jewish/Sephardic/Iberian sister branch, split ca. 400 CE)
R1b-FGC8564 (Iberian sister-branch, split ca. 500 CE; definitely a convert line of ultimately Germanic origin)
J1-FGC5215 (Two Iberian sister-branches [almost certainly converso], split ca. 350 CE)

I would rule out a Romaniote or Mizrahi > K'naan origin for these branches. Non-Rhineland but still Spanish/French/Italian in origin is actually pretty good.

As for:

R1a-Y2619
R1b-A11720
J2a-FGC21085
G-FGC228

The possibility is entirely open (though nothing explicitly supports it, over say, an Italy > Regensburg/Prague route). Let's keep an eye out for Romaniote or Mizrahi connections.

To continue on with this topic

Other Sephardic linked lineages

E-M44>Z17697 via Sephardic-Greece link (Power77 were bring light to your lineage :))

R1b-V88>FGC20980 same concept as above

I would like to post this interesting article

https://www.jewish-eshop.cz/user/documents/upload/Y-DNA_Research_Studies_of_Rabbinical_Lin.pdf

hartaisarlag
01-07-2020, 01:31 AM
To continue on with this topic

Other Sephardic linked lineages

E-M44>Z17697 via Sephardic-Greece link (Power77 were bring light to your lineage :))

R1b-V88>FGC20980 same concept as above

I would like to post this interesting article

https://www.jewish-eshop.cz/user/documents/upload/Y-DNA_Research_Studies_of_Rabbinical_Lin.pdf

I haven't seen that overview! So the Ba'al Shem Tov's Y is a known quantity, eh.

R1b-V88>FGC20980's Ashkenazi-Sephardi divergence ca. 750 suggests an origin in southern France to me, but again, just an inference.

As for E1a-M44>Z17697, the Greek Jew is named Sarfati, which indicates a France > Spain > Greece migration—like any Sephardi with the name "Ashkenazi" who shows up in one of these branches, I take it with a grain of salt.

Principe
01-07-2020, 01:44 AM
I think this conversation is evolving into that. We'll get there.

I actually wanted to tackle the Italy vs. Provence thing first, and I think there are some hints for some lines (for example, divergences ca. 800-900 from Sephardic branches), but there's really no remnant population with 1,200-year-old pedigrees that would allow us to verify this. The Frankfurt project really is a treasure, in that sense. I would love to see as many Italkim as possible tested; if the G-Y12975 Italki line (with a pedigree going back to 14th century Bologna, IIRC) turns out to be basal, then we really have something to say about that branch. J-L816 > J-ZS2728* among Apulian-origin Romaniotes, diverging ca. 900 CE from a large Ashkenazi branch, makes regional Italian speculation possible too in that case.

I think that would be a great idea but its harder to postulate. Similar to G-Y12975, I think J2a-L556 has a strong Rhineland connection it is connected to the Bachrach and Weil Rabbinical lineages as is J1-Y6384 with the Treves lineage, both are present so perhaps both came from Southern France/Provençal region.

StillWater
01-07-2020, 02:31 AM
To continue on with this topic

Other Sephardic linked lineages

E-M44>Z17697 via Sephardic-Greece link (Power77 were bring light to your lineage :))

R1b-V88>FGC20980 same concept as above

I would like to post this interesting article

https://www.jewish-eshop.cz/user/documents/upload/Y-DNA_Research_Studies_of_Rabbinical_Lin.pdf

No one would've ever guessed that Polonsky and the Shpoler Zeyde are Mizrachi.

(Not in the paper): And the real OG, the Lubavitcher Rebbe, might be J-Z482: https://www.familytreedna.com/public/MaHaRaL?iframe=yresults

https://thumbs.gfycat.com/GrandioseTastyKodiakbear-size_restricted.gif

Principe
01-07-2020, 03:04 AM
I haven't seen that overview! So the Ba'al Shem Tov's Y is a known quantity, eh.

R1b-V88>FGC20980's Ashkenazi-Sephardi divergence ca. 750 suggests an origin in southern France to me, but again, just an inference.

As for E1a-M44>Z17697, the Greek Jew is named Sarfati, which indicates a France > Spain > Greece migration—like any Sephardi with the name "Ashkenazi" who shows up in one of these branches, I take it with a grain of salt.

So what are your thoughts then on E-Z17697?

StillWater
01-07-2020, 03:53 AM
I haven't seen that overview! So the Ba'al Shem Tov's Y is a known quantity, eh.

R1b-V88>FGC20980's Ashkenazi-Sephardi divergence ca. 750 suggests an origin in southern France to me, but again, just an inference.

As for E1a-M44>Z17697, the Greek Jew is named Sarfati, which indicates a France > Spain > Greece migration—like any Sephardi with the name "Ashkenazi" who shows up in one of these branches, I take it with a grain of salt.

Are you as skeptical about Ashkenazi lines that come with Sephardic surnames? From what I read, many Sephardic Sarfatis are traditionally descended from Rashi and the name only occurs in Sephardim (maybe also Italkim). Problem is, Rashi only had daughters. Now, if the name is taken in honor of Rashi, then the patrilineal ancestor is likely to be Sephardi. There are also plenty of Moroccan Jews with the surname Sarfati and unlike among Eastern Sephardim, it's hard to find a historical presence of Ashkenazim in North Africa. My guess is that the Sarfatis are Provencal->Sephardi or a very early split from other French Jews, before there was any discernible Ashkenazi YDNA pattern. As tradition goes, it may be in honor of Rashi and thus not affect the YDNA. It could also have been adopted for other reasons. As I mentioned before, there exist Ashkenazim with the surname Ashkenazi, as redundant as it is.

edit: I misread your post. Yes, if it is the sole line among all Ashkenazi lines, it's different. However, the line should really be explored further, because the surname Sarfati isn't as Ashkenazi as it seems (for the aforementioned reasons)

StillWater
01-07-2020, 03:58 AM
Here is a cool YDNA project: https://www.familytreedna.com/public/yichus?iframe=yresults

Has Ashkenazim with traditionally Sephardic surnames (Abarbanel, Doniach etc). Of course, as they're noble names, they could've been adopted from the mother's side, assuming they're not artificial.

hartaisarlag
01-07-2020, 04:51 AM
So what are your thoughts then on E-Z17697?

I don't have many—its Levantine origin seems weirdly open-and-shut considering how odd it is from a deep phylogenetic perspective. But so far, no structurally conclusive proof of a broader Jewish connection. Sarfati is worth keeping an eye on.


Are you as skeptical about Ashkenazi lines that come with Sephardic surnames? From what I read, many Sephardic Sarfatis are traditionally descended from Rashi and the name only occurs in Sephardim (maybe also Italkim). Problem is, Rashi only had daughters. Now, if the name is taken in honor of Rashi, then the patrilineal ancestor is likely to be Sephardi. There are also plenty of Moroccan Jews with the surname Sarfati and unlike among Eastern Sephardim, it's hard to find a historical presence of Ashkenazim in North Africa. My guess is that the Sarfatis are Provencal->Sephardi or a very early split from other French Jews, before there was any discernible Ashkenazi YDNA pattern. As tradition goes, it may be in honor of Rashi and thus not affect the YDNA. It could also have been adopted for other reasons. As I mentioned before, there exist Ashkenazim with the surname Ashkenazi, as redundant as it is.

edit: I misread your post. Yes, if it is the sole line among all Ashkenazi lines, it's different. However, the line should really be explored further, because the surname Sarfati isn't as Ashkenazi as it seems (for the aforementioned reasons)

So the real question is, is the Sarfati line basal to the branch or not? If it is, which the FTDNA Haplotree suggests is a possibility, then StillWater's divergence in Provence theory is a perfect fit. And I'll bet the updated TMRCA would fall somewhere around 750 to 900. I was unaware of the Rashi tradition, but that's really interesting. The source I read last year on the Sarfati name in North Africa didn't mention it or Provence specifically, but seemed pretty sure it was an accurate reflection of France > (Spain >) North Africa migration history.

My take on Ashkenazim with Sephardi surnames is: get NGS Y chromosome-tested. If they are nested deep within Ashkenazi branches, then it's probably fictitious; if they're basal or fall into subclades unheard of among Ashkenazim, then there's something to it. I'm sure there are at least a few cases we could look to.

Piquerobi
01-07-2020, 12:49 PM
Here is an interesting project, that of the Jews of Frankfurt:


My primary mission is to find people who are direct line offspring of the families of the Frankfurt region, (which I've extended to include Worms, Mainz, Metz and other nearby towns). Frankfurt is special. Thanks to the database known as "Ele Toldot", it is possible to reconstruct family trees for all Jews who lived there following the Black Plague in 1349. Neighboring Worms and Mainz were very early Rabbinical centers, and families from early Worms were the ancestors for many of the later Jews of Frankfurt; Guggenheim, Wertheim, Oppenheim, Bach(a)rach, ... just to name a few. To this end, I have already entered the Ele Toldot records for all of the early families of Frankfurt as well as from other sources into my database. Along with my other administrators, we are very happy to help all of our members (of course free of charge) to learn more about their family. I also am willing to create haplogroup-specific gedcoms for each cluster, and I am quite willing to share this information with the members of my in my project as well.
https://www.familytreedna.com/groups/jews-of-frankfurt/about/goals
https://www.familytreedna.com/public/Jews_of_Frankfurt?iframe=yresults

An informative site on the Jewish quarter of Frankfurt:

http://www.judengasse.de/ehtml/page812.htm

StillWater
01-07-2020, 06:15 PM
I don't have many—its Levantine origin seems weirdly open-and-shut considering how odd it is from a deep phylogenetic perspective. But so far, no structurally conclusive proof of a broader Jewish connection. Sarfati is worth keeping an eye on.



So the real question is, is the Sarfati line basal to the branch or not? If it is, which the FTDNA Haplotree suggests is a possibility, then StillWater's divergence in Provence theory is a perfect fit. And I'll bet the updated TMRCA would fall somewhere around 750 to 900. I was unaware of the Rashi tradition, but that's really interesting. The source I read last year on the Sarfati name in North Africa didn't mention it or Provence specifically, but seemed pretty sure it was an accurate reflection of France > (Spain >) North Africa migration history.

My take on Ashkenazim with Sephardi surnames is: get NGS Y chromosome-tested. If they are nested deep within Ashkenazi branches, then it's probably fictitious; if they're basal or fall into subclades unheard of among Ashkenazim, then there's something to it. I'm sure there are at least a few cases we could look to.

If you look at the Frankfurt Project, you'll find the following header: "J1: TREVES (ASHKENAZI/TSARFATI) 16th C Italy". This is another line descended from one of Rashi's daughters. The surname of the line is Treves (after Rashi's birth city of Troyes and precisely because Rashi was born there, at least according to tradition. I've met 2 Treveses irl; both claimed descent from Rashi on their paternal line) and this family also associates it with being "Tsarfati".

Also, just found this: https://www.igenea.com/en/surname-projects/a/ashkenazi-treves-6170

Also, see this: https://www.jewishvirtuallibrary.org/treves

http://freepages.rootsweb.com/~prohel/genealogy/names/loew/spira/spira_2.html

Pylsteen
01-07-2020, 06:53 PM
I have two Sarfati families in my tree, one very likely from Portugal, from which also Samuel Sarphati (https://en.wikipedia.org/wiki/Samuel_Sarphati) descends, and another from Northern Italy, probably originally from 15th/16th century Provence. They became part of the Amsterdam Sephardic community.
This suggests to me that the Sephardic community in Amsterdam was itself, although largely made up of Spanish, Portugese, Italian and Ottoman Sephardim, already mixed with Jews with a French or possibly Ashkenazi origin before it started to mix with the Ashkenazi community during the 19th century.

The Ele Toldot is a great source for Frankfurt Jews; I could make a tree of one of my Ashkenazi lines; unfortunately, I could not place my own ancestor in this family, because she (and probably her father too) did not die in Frankfurt. In the late 18th century she first went to Paris and with her husband from Strasbourg she settled in Amsterdam.

Principe
01-07-2020, 10:29 PM
If you look at the Frankfurt Project, you'll find the following header: "J1: TREVES (ASHKENAZI/TSARFATI) 16th C Italy". This is another line descended from one of Rashi's daughters. The surname of the line is Treves (after Rashi's birth city of Troyes and precisely because Rashi was born there, at least according to tradition. I've met 2 Treveses irl; both claimed descent from Rashi on their paternal line) and this family also associates it with being "Tsarfati".

Also, just found this: https://www.igenea.com/en/surname-projects/a/ashkenazi-treves-6170

Also, see this: https://www.jewishvirtuallibrary.org/treves

http://freepages.rootsweb.com/~prohel/genealogy/names/loew/spira/spira_2.html

In all honesty it would probably be virtually impossible to know Rashi’s Y line, but we actually have a good idea of what it could have been, essentially one of those Rhineland Y that me and Hartaisarlag were discussing about.

StillWater
01-07-2020, 10:42 PM
In all honesty it would probably be virtually impossible to know Rashi’s Y line, but we actually have a good idea of what it could have been, essentially one of those Rhineland Y that me and Hartaisarlag were discussing about.

By oral history, I am descended from Rabbeinu Tam (Rashi's grandson) on one branch. I would not be shocked if Rashi did not have a typical Rhine YDNA, as there is a theory that his ancestors moved north from Lunel (Provence).

StillWater
01-07-2020, 10:49 PM
You guys seen this cartoon?


https://www.youtube.com/watch?v=fSLKfzrIXM8

In it, French Jewry is said to have arrived from Babylon.

eolien
01-08-2020, 09:45 AM
I haven't seen that overview! So the Ba'al Shem Tov's Y is a known quantity, eh.

R1b-V88>FGC20980's Ashkenazi-Sephardi divergence ca. 750 suggests an origin in southern France to me, but again, just an inference.

As for E1a-M44>Z17697, the Greek Jew is named Sarfati, which indicates a France > Spain > Greece migration—like any Sephardi with the name "Ashkenazi" who shows up in one of these branches, I take it with a grain of salt.

There is no historical record that Jews of Tsarfat (northern france) moved to Spain in big numbers. Some Rabbis etc might have moved but physically it was almost impossible. Therefore when they were expelled they settled in Burgundy and Germany. In fact there is evidence for their resettlement to Cologne and Trier. I emphasized this fact in my previous posts in our discussion about origins of the Rhein community and Yiddish.

Likewise the jews of Provence moved to italy (except pope's jews) because Tsarfat and Iberia were already empty of Jews at that time. We cannot exclude that some prominent jews decided to migrate from Provence to Catalunia in the 15th century or before, but this was not a general phenomenon. But we know is however that many Spanish jews escaped to Provence beforehand during persecutions.

Once you were in the Northern italy in the 16th century you could move easily back and forth between there and Ottoman lands because the whole trade of Tuscany, Venice etc was based in the Levant. Moreover the Ashkenazi communities had limited space and restricted residence rights that did not allow them to accept large number of refugees. For example Frankfurt ghetto could only have 200 families and carry 15 marriages per year.

hartaisarlag
01-08-2020, 06:23 PM
There is no historical record that Jews of Tsarfat (northern france) moved to Spain in big numbers. Some Rabbis etc might have moved but physically it was almost impossible. Therefore when they were expelled they settled in Burgundy and Germany. In fact there is evidence for their resettlement to Cologne and Trier. I emphasized this fact in my previous posts in our discussion about origins of the Rhein community and Yiddish.

Likewise the jews of Provence moved to italy (except pope's jews) because Tsarfat and Iberia were already empty of Jews at that time. We cannot exclude that some prominent jews decided to migrate from Provence to Catalunia in the 15th century or before, but this was not a general phenomenon. But we know is however that many Spanish jews escapted to Provence beforehand during persecutions.

Once you were in the Northern italy in the 16th century you could moved easily back and forth between there and Ottoman lands because the whole trade of Tuscany, Venice etc was based in the Levant. Moreover the Ashkenazi communities had limited space and restricted residence rights that did not allow them to accept large number of refugees. For example Frankfurt ghetto could only have 200 families and carry 15 marriages per year.

So as StillWater was suggesting, maybe the name Sarfati in that case is a reflection of moves from Provence/Septimania, rather than Tsarfat proper.

But the truth is, I can't find great sources on the direction of migrations, pre-1000, between Jewish communities in Italy, Provence, and Spain. My assumption from parsimony has been that branches with an Ashkenazi-Sephardi divergence between 750 and 900 might've split in Provence, but I'm not claiming to have hard evidence. In at least one of these cases (R1b-V3476), it's clear that the ultimate origin was Iberia, but the fact that two sister-branches under it are Ashkenazi suggests that their proximate origin was somewhere that directly supplied migrations to early Ashkenaz.

hartaisarlag
01-08-2020, 07:03 PM
Your point also gets at an issue we haven’t discussed in this thread, or discussed enough anywhere, which is: what was the primary origin (in a proximate geographical sense, not in a population components sense) of early medieval Iberian Jews?

What share of their origins go back to late classical Italy and Provence, vs. North Africa? My understanding is that the only strong evidence for classical continuity of Jewish settlement is in Catalonia, but I could be wrong. Besides the fact that this is an important issue that’s actually a lot murkier than the Ashkenazi origins question, selfishly, it has direct implications for the history of my branch, E-Y6923, which has been found in both North African Jews and Iberians of almost certain converso descent.

Another angle on this question: to what degree does the base of the Sephardi population (ca. 1492; we obviously run into other issues w/Romaniote and Ashkenazi contributions in the Balkans) derive from the sources that gave rise to medieval Ashkenazim? For a decade, the autosomal evidence has led us to assume ‘a lot’—but we can’t say on that basis whether the source populations of pre-Ashkenaz and pre-Sepharad mostly diverged in classical times, or at some later point.

What’s clear from this thread’s investigation is that there’s plenty of evidence for overlap, probably at multiple post-classical time points. But our Ashkenaz-centricity shouldn’t lead us to conclude that just because a large % of Ashkenazi lines are related to Sephardim, that a large % of Sephardi lines are related to Ashkenazim. Squares and rectangles. After all, Ashkenazim are a bottlenecked population; a small northerly subset of western Jewish diversity.

John Doe
01-09-2020, 04:38 PM
You guys seen this cartoon?


https://www.youtube.com/watch?v=fSLKfzrIXM8

In it, French Jewry is said to have arrived from Babylon.

Wow! I remember this movie, they showed it to us at school years ago. How quickly time passes lol

Cascio
01-09-2020, 04:57 PM
Delete

Cascio
01-09-2020, 04:59 PM
Your point also gets at an issue we haven’t discussed in this thread, or discussed enough anywhere, which is: what was the primary origin (in a proximate geographical sense, not in a population components sense) of early medieval Iberian Jews?

What share of their origins go back to late classical Italy and Provence, vs. North Africa? My understanding is that the only strong evidence for classical continuity of Jewish settlement is in Catalonia, but I could be wrong. Besides the fact that this is an important issue that’s actually a lot murkier than the Ashkenazi origins question, selfishly, it has direct implications for the history of my branch, E-Y6923, which has been found in both North African Jews and Iberians of almost certain converso descent.

Another angle on this question: to what degree does the base of the Sephardi population (ca. 1492; we obviously run into other issues w/Romaniote and Ashkenazi contributions in the Balkans) derive from the sources that gave rise to medieval Ashkenazim? For a decade, the autosomal evidence has led us to assume ‘a lot’—but we can’t say on that basis whether the source populations of pre-Ashkenaz and pre-Sepharad mostly diverged in classical times, or at some later point.

What’s clear from this thread’s investigation is that there’s plenty of evidence for overlap, probably at multiple post-classical time points. But our Ashkenaz-centricity shouldn’t lead us to conclude that just because a large % of Ashkenazi lines are related to Sephardim, that a large % of Sephardi lines are related to Ashkenazim. Squares and rectangles. After all, Ashkenazim are a bottlenecked population; a small northerly subset of western Jewish diversity.

Were the Provencal Jews Sephardim or Ashkenazim or sui generis like Italkim?

Dewsloth
01-09-2020, 08:22 PM
To continue the point further the Rhineland massacres would have potentially created scenarios where some branches could have had older TMRCA’s but these branches were killed off and that’s why younger TMRCA’s exist.

But you see here’s where it gets interesting all the sources of R-Y2619 seem to be coming from the East via Bavaria and Prague, it also seems the Levite family of Epstein seems to have been R1b-L2>L208, so a Bavaria/Bohemia/Elbe origin for Y2619 branch in terms of Community makes sense. In the case of E-Y6938 there are no branches in this community and I think the diversity of subranching would suggest a Rhineland origin is less likely so similarly a Bavaria/Bohemia/Elbe origin is likely for E-Y6938.

I think in the case of R-A11720 and J2a-FGC21085 they follow a similar pattern to Y6938 and Y2619 where they diversified old enough to be also from the Bavaria/Bohemia/Elbe communities as well.

G-FGC228 might be a prime example of how the Rhineland massacres impacted the TMRCA’s or lineage appearance, so I see it possible that FGC228 could have been part of the original Rhineland communities or it could have have come from the Eastern Trio, in all likelihood this branch pre Ashkenazi should have came from either North-Central Italian or Provençal/Southern France Jewish communities.

As for R-FGC8564 and J1-FGC5215, in all honesty yes a later Sephardic arrival makes sense, I know there is no explicit Sephardic origin for the Ivanhoe clade but the Spanish lines appear pretty close to the TMRCA of the Ashkenazim line, so at least an Eastern Germanic origin pre Jewish conversion is almost a given, so if not Spain, Southern France or Italy are likely as well, likely either Visigoth or Suebi in origin, Vandals are likely eliminated because they were already out by 500. The TMRCA of FGC8564 fits perfectly with either or.

I just wrote this in the Big Y thread:

"My fourth Y match joined the DF19 group today, after a few months of me begging. So we.have 4 families, different surnames and not considered STR matches with each other by FTDNA, all with MDKAs in Hesse dating back to the 1600s-1800s. And one clade level up, a fifth family in northern Germany. I'm still thinking they stem from some Elbe or Rhine Germanic migration, but so many Germanic and northern Celtic tribes occupied the same space at different times, there's no hard evidence of anything in particular"

^^The fourth Y match is Jewish, and his line came to the US in the 1800s.
The estimate that the DF19 group admins had for the southern Hesse/Rhine trio was around 400AD, with the older shared clade northern one in Stendal (right next to the Elbe, but MDKA only back to the 1800s) forming some time earlier.

We're all in this block:

https://www.ytree.net/DisplayTree.php?blockID=3410&star=false

My own Jewish non-Y ancestors also have MDKA in the Rhineland, but I don't know what their Y line is yet.

Claudio
01-10-2020, 08:25 PM
Your point also gets at an issue we haven’t discussed in this thread, or discussed enough anywhere, which is: what was the primary origin (in a proximate geographical sense, not in a population components sense) of early medieval Iberian Jews?

What share of their origins go back to late classical Italy and Provence, vs. North Africa? My understanding is that the only strong evidence for classical continuity of Jewish settlement is in Catalonia, but I could be wrong. Besides the fact that this is an important issue that’s actually a lot murkier than the Ashkenazi origins question, selfishly, it has direct implications for the history of my branch, E-Y6923, which has been found in both North African Jews and Iberians of almost certain converso descent.

Another angle on this question: to what degree does the base of the Sephardi population (ca. 1492; we obviously run into other issues w/Romaniote and Ashkenazi contributions in the Balkans) derive from the sources that gave rise to medieval Ashkenazim? For a decade, the autosomal evidence has led us to assume ‘a lot’—but we can’t say on that basis whether the source populations of pre-Ashkenaz and pre-Sepharad mostly diverged in classical times, or at some later point.

What’s clear from this thread’s investigation is that there’s plenty of evidence for overlap, probably at multiple post-classical time points. But our Ashkenaz-centricity shouldn’t lead us to conclude that just because a large % of Ashkenazi lines are related to Sephardim, that a large % of Sephardi lines are related to Ashkenazim. Squares and rectangles. After all, Ashkenazim are a bottlenecked population; a small northerly subset of western Jewish diversity.

I’ve been watching quite a few vids by Dr.Henry Abramson,these two vids if you have the time to look up on YouTube were super interesting:
35793
He’s quite evident in his chronology establishing there is No evidence of Jewish communities in Spain prior to 1st century AD,He has the Jews of 200AD - 600AD as First Latin and Greek Speakers then Solely Latin but knowing Hebrew and using in ritual sense both orally and written.
He never brings genetics into his discussions so watching his videos from the outside with genetics in the peripheral vision you can make some observations.
It would seem under the Visigoths from reading in between the lines to what was banned,there was a problem of interaction between Christian and Jew as well as a problem of mixed marriages,which would suggest iberian admixture addded at this point,but what’s really interesting in his second video regarding Spain under Islam is his evident observation that Spain being a tolerant society with vast fertile land,with opportunities and a pleasant climate that this attracted Jewish migrants from all over the Islamic world at first from Tunisia then progressively from Libya,Egypt then Levant and Mesopotamia.
Separately reading up on the history of the Jews of Tunisia,myths aside evidence points to them being descendants of Cyrenaican,Alexandrian and Roman Jewish exiles,would seem Kairouan in Tunisia was a major Jewish pole in both Roman,vandal and early Islamic eras.
It suggests to me Sephardic Jews of Spain prior 15th century expulsions already had North African Admixture and owing there ancestry to South euro admixed Roman/Visigoth era Jews mixed with North African and Mesopotamian Jews all carrying various degrees of Levantine Ancestry.
It’s a shame that ( as far as I can see) Abramson has not posted a video solely on the history of Ashkenazi Jews as would really like his take,but bringing back the old subject of Ashkenazim also seeming to have elevated berber North African admixture (as if they share an ethnogenesis with Sephardim) could a combination of shared Sephardic and perhaps Tunisian Jewry from Roman North Africa ( who in turn owe there Ancestry to Cyrenaican,Alexandrian and Judean Levantine Jews) migrating up through Roman Tunisia Italian peninsula be a seed population for both Sephardim and Ashkenazim resulting in the shared berber component to lesser extents?

passenger
01-10-2020, 09:00 PM
I’ve been watching quite a few vids by Dr.Henry Abramson,these two vids if you have the time to look up on YouTube were super interesting:
35793
He’s quite evident in his chronology establishing there is No evidence of Jewish communities in Spain prior to 1st century AD,He has the Jews of 200AD - 600AD as First Latin and Greek Speakers then Solely Latin but knowing Hebrew and using in ritual sense both orally and written.
He never brings genetics into his discussions so watching his videos from the outside with genetics in the peripheral vision you can make some observations.
It would seem under the Visigoths from reading in between the lines to what was banned,there was a problem of interaction between Christian and Jew as well as a problem of mixed marriages,which would suggest iberian admixture addded at this point,but what’s really interesting in his second video regarding Spain under Islam is his evident observation that Spain being a tolerant society with vast fertile land,with opportunities and a pleasant climate that this attracted Jewish migrants from all over the Islamic world at first from Tunisia then progressively from Libya,Egypt then Levant and Mesopotamia.
Separately reading up on the history of the Jews of Tunisia,myths aside evidence points to them being descendants of Cyrenaican,Alexandrian and Roman Jewish exiles,would seem Kairouan in Tunisia was a major Jewish pole in both Roman,vandal and early Islamic eras.
It suggests to me Sephardic Jews of Spain prior 15th century expulsions already had North African Admixture and owing there ancestry to South euro admixed Roman/Visigoth era Jews mixed with North African and Mesopotamian Jews all carrying various degrees of Levantine Ancestry.
It’s a shame that ( as far as I can see) Abramson has not posted a video solely on the history of Ashkenazi Jews as would really like his take,but bringing back the old subject of Ashkenazim also seeming to have elevated berber North African admixture (as if they share an ethnogenesis with Sephardim) could a combination of shared Sephardic and perhaps Tunisian Jewry from Roman North Africa ( who in turn owe there Ancestry to Cyrenaican,Alexandrian and Judean Levantine Jews) migrating up through Roman Tunisia Italian peninsula be a seed population for both Sephardim and Ashkenazim resulting in the shared berber component to lesser extents?

Love Henry Abramson! But in terms of early Iberian Jewish history I think Benjamin Gampel does a more thorough job in his series, which I may have posted before. This is the first episode: https://www.youtube.com/watch?v=GRZkxypF05M&list=PL7Yaf7nQHP3CD-jtibqiM_NOOAll4s6JA

I've often seen it cited that medieval Iberia, particularly Al-Andalus, was a magnet for Jews from almost every part of the diaspora, but unfortunately the paper trail tends to dead-end with references from late-19th and early 20th century historians who were not always the best at backing up their statements with reliable statistics. I have yet to find a detailed description of medieval Jewish migration to Iberia, though it seems very plausible that it occurred. If anyone has any info, let me know!

And yes, it seems logical that Roman North Africa would play a role in the Berber ancestry shared by Western Jews.

ADW_1981
01-10-2020, 10:05 PM
I'm not sure exactly how the results were tabulated, based on predicted YSTR or deep SNP tests, but one resource that has quite a few legitimate R1b members is this project: https://www.familytreedna.com/groups/jewish-r1b/dna-results

Obviously you'd need to do some trimming of the results, but the V88 and M269(xL23) groups are quite solid.

Also, I have a hunch that the L21+ group is Iberian, not British, as the Kimhi line shows a west to east distribution. I'm just not sure if it has the same terminal SNP that you have listed as the core L21+ group though.

StillWater
01-13-2020, 11:12 PM
You guys should look at: https://www.familytreedna.com/public/Luria?iframe=yresults

This is a very old family, and connections (genealogical and phylogenic) can be made to Italian Jews, some with a deep pedigree in Italy. It's one of the oldest Jewish families in Europe. You'll notice that Italian Jews appear in some of those clades, though in some cases, the clades have a distant MRCA.

while the claims made here may no longer stand up, this is a good read to understand their significance: https://www.jta.org/2004/10/24/lifestyle/the-oldest-family-in-the-world

This family is likely crucial to solving the entire puzzle.

hartaisarlag
01-15-2020, 12:19 AM
Of major note: the newest YFull update, currently underway, has one unlabeled individual splitting J-Y3088* from J-Z18271. I want to figure out who this is.

EDIT: A Brazilian with no known Jewish roots.

hartaisarlag
01-16-2020, 08:14 PM
Of major note: the newest YFull update, currently underway, has one unlabeled individual splitting J-Y3088* from J-Z18271. I want to figure out who this is.

EDIT: A Brazilian with no known Jewish roots.

Alright, looking forward to others’ takes, but something that occurred to me: whether we trust the ca. 1400 BCE divergence date or not, it’s very hard to think of a mechanism by which the most basal lineage in a pan-Jewish lineage ends up in Sephardim, and thence, gentile Portuguese.

I’m tempted to look elsewhere. Suppose J-Z18271 is the firmly Israelite lineage, nested within the wider J-Y3088/ZS222. The basal Brazilian could be the descendant of a Phoenician, or even of an Arab / Arabized Levantine.

StillWater
01-17-2020, 12:33 AM
Alright, looking forward to others’ takes, but something that occurred to me: whether we trust the ca. 1400 BCE divergence date or not, it’s very hard to think of a mechanism by which the most basal lineage in a pan-Jewish lineage ends up in Sephardim, and thence, gentile Portuguese.

I’m tempted to look elsewhere. Suppose J-Z18271 is the firmly Israelite lineage, nested within the wider J-Y3088/ZS222. The basal Brazilian could be the descendant of a Phoenician, or even of an Arab / Arabized Levantine.

Why must a branch that leaves home develop more mutations than the one that remains? I don't see the conflict. Of course, we then still expect a high chance of Arabs etc in the basal clade as well.

hartaisarlag
01-17-2020, 12:46 AM
Why must a branch that leaves home develop more mutations than the one that remains? I don't see the conflict. Of course, we then still expect a high chance of Arabs etc in the basal clade as well.

It's not that the J-Y3088* individual has fewer mutations in toto, it's that he doesn't share a whole boatload of mutations common to all Ashkenazi, Sephardi, Italian, Mizrahi, and Yemenite Jews in the branch. Whether this basal individual diverged from J-Z18271 ca. 1400 BCE, as claimed by YFull, or 500 years before the Behar et al. 2017 estimate of 6th century BCE, this is the period in which we'd expect a proto-Israelite gene pool to be forming. Two lines diverging during this period could either represent a split between various Northwest/Central Semitic populations, or a very early branching among Israelites which somehow manifested very asymmetrically in the diaspora 1,000-1,500 years later.

hartaisarlag
01-20-2020, 07:59 PM
Worth bringing to people's attention that the 12th-largest lineage, R-A11720, which is the matter of some dispute, just got even more interesting. Checked YFull, and its sister branch, R-BY16680 (split dated to around 2600 BCE), now has its first Middle Eastern member, a Kuwaiti. FTDNA assigns this Kuwaiti to R-FT44409, alongside an Italian and a Romanian. Not sure how to make sense of this yet, but the temptation to chalk this up to a simple Balkan story is getting harder to sustain.

hartaisarlag
01-22-2020, 01:07 AM
I either missed one of these, or the latest update to YFull has newly cast it in relief.

#39, J2a-FGC4975, seems to have been the subject of a recent upload drive by the Avotaynu project, and apparently has a TMRCA of 0. I believe JewishDNA.net incorrectly labels only one of its two sub-branches, J2a-L254, as an Ashkenazi branch.

This one is a decent candidate for an Ashkenazi J2a branch with a European Mediterranean origin, though it's hard to say definitively.

I also somehow missed the fact that J2a-FGC4975 appears to be the largest J2a Kohen branch—and what it has in common with J1-Z18271 and J2b-Y33795 is a very early TMRCA. As far as I can tell, these are the 3 largest Kohen branches.

hartaisarlag
01-22-2020, 09:13 PM
To any who are interested: a New Mexican with roots going back to 17th century Ciudad Juárez/El Paso has tested positive for Y6923, but negative for both Y6938 and Y102667. The structure of this subclade just got a lot more interesting. The way I see it, there are 4 active possibilities, which we'll hopefully be able to test using YFull:

1) This new individual forms a branch coequal with Y6938 and Y102667
2) He forms a basal branch in opposition to a branch uniting Y6938 and Y102667 (there might be circumstantial proof for this in the fact that he has no listed matches closer than Y12)
3) He forms a basal branch joining him with Y6938 in opposition to Y102667
4) He forms a basal branch joining him with Y102667 in opposition to Y6938

Whatever the case, this makes it likelier that Iberia plays an important role in the story—just how important is to be determined.

ADDENDUM: This means that for the other Libyan Jewish, Algerian Jewish, Tunisian Jewish, Peruvian, and Sicilian individuals who are either SNP-confirmed for E-Y6923 or predicted members of the clade, we can no longer assume membership in E-Y102667. Anything is possible.

VytautusofAukstaitija
01-23-2020, 02:29 PM
I’ve been watching quite a few vids by Dr.Henry Abramson,these two vids if you have the time to look up on YouTube were super interesting:
35793
He’s quite evident in his chronology establishing there is No evidence of Jewish communities in Spain prior to 1st century AD,He has the Jews of 200AD - 600AD as First Latin and Greek Speakers then Solely Latin but knowing Hebrew and using in ritual sense both orally and written.
He never brings genetics into his discussions so watching his videos from the outside with genetics in the peripheral vision you can make some observations.
It would seem under the Visigoths from reading in between the lines to what was banned,there was a problem of interaction between Christian and Jew as well as a problem of mixed marriages,which would suggest iberian admixture addded at this point,but what’s really interesting in his second video regarding Spain under Islam is his evident observation that Spain being a tolerant society with vast fertile land,with opportunities and a pleasant climate that this attracted Jewish migrants from all over the Islamic world at first from Tunisia then progressively from Libya,Egypt then Levant and Mesopotamia.
Separately reading up on the history of the Jews of Tunisia,myths aside evidence points to them being descendants of Cyrenaican,Alexandrian and Roman Jewish exiles,would seem Kairouan in Tunisia was a major Jewish pole in both Roman,vandal and early Islamic eras.
It suggests to me Sephardic Jews of Spain prior 15th century expulsions already had North African Admixture and owing there ancestry to South euro admixed Roman/Visigoth era Jews mixed with North African and Mesopotamian Jews all carrying various degrees of Levantine Ancestry.
It’s a shame that ( as far as I can see) Abramson has not posted a video solely on the history of Ashkenazi Jews as would really like his take,but bringing back the old subject of Ashkenazim also seeming to have elevated berber North African admixture (as if they share an ethnogenesis with Sephardim) could a combination of shared Sephardic and perhaps Tunisian Jewry from Roman North Africa ( who in turn owe there Ancestry to Cyrenaican,Alexandrian and Judean Levantine Jews) migrating up through Roman Tunisia Italian peninsula be a seed population for both Sephardim and Ashkenazim resulting in the shared berber component to lesser extents?

I don't know if this means much, but in the later period before the conquests the Visigoths clamped down on Iberian Jewry but the Jewish communities on the Visigothic marches may have had a protected status of sorts.

The situation in southern France was quite different, certain lords relied on some degree of Jewish manpower for their military forces, so they presumably lived quite a different experience from the Jews under Rodrigo or Witiza.

Alot of the urban garrison forces in Iberia were Jewish as well during the initial conquest of Tariq and Musa.

Basically the picture look like Jewish communities in Iberia and southern France under certain periods had wielded significant military power, and gained them favorable status. I think this was stronger in Al-Andalus compared to the rather brief experience in parts of southern France.

Agamemnon
01-25-2020, 05:30 PM
Alright, looking forward to others’ takes, but something that occurred to me: whether we trust the ca. 1400 BCE divergence date or not, it’s very hard to think of a mechanism by which the most basal lineage in a pan-Jewish lineage ends up in Sephardim, and thence, gentile Portuguese.

I’m tempted to look elsewhere. Suppose J-Z18271 is the firmly Israelite lineage, nested within the wider J-Y3088/ZS222. The basal Brazilian could be the descendant of a Phoenician, or even of an Arab / Arabized Levantine.

Been away for some time (busy), but I did notice the update and I found it both encouraging and discouraging. I (and others familiar with ZS241) have known for several years that YFull's tree was severely flawed both in terms of dating and leveling of SNP equivalents. The recent update is a big step in the right direction, with Y3088 as a ZS222 equivalent (something I suspected for some time), in effect placing Z18271's TMRCA in the Iron Age and its parent in an LBA context. Another prediction I want to make is that ZS227 is also going to be a Y3088 equivalent, and that the age of the branch under which it is now subsumed (S20075) will also come down.

Now regarding the line of reasoning you put forward, I can understand why a Jewish origin would seem problematic at first, but ultimately that's the most plausible explanation... And not necessarily because the area from which the Brazilian individual hails is known to be a crypto-Jewish reservoir.

There is little doubt in my mind that ZS222 (Y3088) is going to be an essentially Israelite lineage, one might even call it "Proto-Israelite". As I said above, its TMRCA falls in an LBA-IA I context, which is the time frame in which the Israelites emerged. This is roughly comparable to J2a-Y15222 and other Jewish (and Samaritan) lineages, possibly including E-Y6926. Furthermore, it's unlikely we'll ever find Z18271 in an Iron Age Judahite or Israelite context, I cannot stress this enough but the way things are looking right now Z18271 is going to be strictly associated with the priestly class (which was much more restricted all the way until the IA II-C), the only ZS241 we'll find in such contexts is going to be ZS222. The reason for this is that the lineages we see in present-day Jews mostly reflect a subset of Judean-society, most notably the higher stratum, there are of course possible exceptions... And we're probably dealing with one here. I'd argue that things were reversed for ZS222, and that we only get the impression that its presence in Jews/Israelites is limited to the priests because of later selection for Kohanic pedigrees (and so Z18271-rich), prior to the Babylonian exile (and possibly even after that) at least odds are that the bulk of the ZS222 among Jews was xZ18271. I'd also expect to find some of these lineages in a minority of Jews, and so it wouldn't be surprising if they fared better among descendants of crypto-Jews or Jews who "opted out" of Jewish society thanks to persecution.

Even more relevant in my view is the fact that ZS222 is a poor candidate for a Phoenician lineage, the only reasonably convincing contender under ZS241 for a Phoenician association is BY143137. Again, I'd be amazed if it turned out that ZS222 was anything other than a quintessentially Israelite lineage (or, if we're going to be precise, "Highland Canaanite"). Depending on whether or not there is any truth to the Aaronid pedigree (which should not be discarded out of hand as a convincing argument can be made in its favour), the most we can say is that if an Egyptian background among the so-called "Asiatics" (Aamu) were to be rejected ZS222's origins would most likely lie in one of the city-states of southern Canaan (possibly even Jerusalem), in an area roughly corresponding to the green-shaded area we see in the following map depicting early Israelite settlement:


https://i.imgur.com/L17TtaE.png

Beyond that, it's hard to say what ZS222's ultimate provenance is. ZS227 (immediately upstream) could either be Western Amorite or Early Canaanite, the extent to which these can be fully dissociated is somewhat doubtful as well. In any case, a Phoenician origin doesn't really strike me as a plausible alternative. One thing to keep in mind however is that ZS222 might have existed simultaneously amongst the Israelites, Ammonites and Moabites (again "Highland Canaanite").

hartaisarlag
01-28-2020, 08:12 PM
Been away for some time (busy), but I did notice the update and I found it both encouraging and discouraging. I (and others familiar with ZS241) have known for several years that YFull's tree was severely flawed both in terms of dating and leveling of SNP equivalents. The recent update is a big step in the right direction, with Y3088 as a ZS222 equivalent (something I suspected for some time), in effect placing Z18271's TMRCA in the Iron Age and its parent in an LBA context. Another prediction I want to make is that ZS227 is also going to be a Y3088 equivalent, and that the age of the branch under which it is now subsumed (S20075) will also come down.

Now regarding the line of reasoning you put forward, I can understand why a Jewish origin would seem problematic at first, but ultimately that's the most plausible explanation... And not necessarily because the area from which the Brazilian individual hails is known to be a crypto-Jewish reservoir.

There is little doubt in my mind that ZS222 (Y3088) is going to be an essentially Israelite lineage, one might even call it "Proto-Israelite". As I said above, its TMRCA falls in an LBA-IA I context, which is the time frame in which the Israelites emerged. This is roughly comparable to J2a-Y15222 and other Jewish (and Samaritan) lineages, possibly including E-Y6926. Furthermore, it's unlikely we'll ever find Z18271 in an Iron Age Judahite or Israelite context, I cannot stress this enough but the way things are looking right now Z18271 is going to be strictly associated with the priestly class (which was much more restricted all the way until the IA II-C), the only ZS241 we'll find in such contexts is going to be ZS222. The reason for this is that the lineages we see in present-day Jews mostly reflect a subset of Judean-society, most notably the higher stratum, there are of course possible exceptions... And we're probably dealing with one here. I'd argue that things were reversed for ZS222, and that we only get the impression that its presence in Jews/Israelites is limited to the priests because of later selection for Kohanic pedigrees (and so Z18271-rich), prior to the Babylonian exile (and possibly even after that) at least odds are that the bulk of the ZS222 among Jews was xZ18271. I'd also expect to find some of these lineages in a minority of Jews, and so it wouldn't be surprising if they fared better among descendants of crypto-Jews or Jews who "opted out" of Jewish society thanks to persecution.

Even more relevant in my view is the fact that ZS222 is a poor candidate for a Phoenician lineage, the only reasonably convincing contender under ZS241 for a Phoenician association is BY143137. Again, I'd be amazed if it turned out that ZS222 was anything other than a quintessentially Israelite lineage (or, if we're going to be precise, "Highland Canaanite"). Depending on whether or not there is any truth to the Aaronid pedigree (which should not be discarded out of hand as a convincing argument can be made in its favour), the most we can say is that if an Egyptian background among the so-called "Asiatics" (Aamu) were to be rejected ZS222's origins would most likely lie in one of the city-states of southern Canaan (possibly even Jerusalem), in an area roughly corresponding to the green-shaded area we see in the following map depicting early Israelite settlement:


https://i.imgur.com/L17TtaE.png

Beyond that, it's hard to say what ZS222's ultimate provenance is. ZS227 (immediately upstream) could either be Western Amorite or Early Canaanite, the extent to which these can be fully dissociated is somewhat doubtful as well. In any case, a Phoenician origin doesn't really strike me as a plausible alternative. One thing to keep in mind however is that ZS222 might have existed simultaneously amongst the Israelites, Ammonites and Moabites (again "Highland Canaanite").

Touché. Unfortunately we're impoverished re: NGS Y-tested Levantines, so making any definitive sub-regional associations is hard. A few smaller Ashkenazi branches have definitive Palestinian or Lebanese links within the last 2,000-4,000 years, but without a concerted effort to find more, I expect this part of the mystery will take a long time to solve, longer than the associated proto-historic or classical/post-classical ones.

artemv
01-30-2020, 05:21 AM
I (and others familiar with ZS241) have known for several years that YFull's tree was severely flawed both in terms of dating and leveling of SNP equivalents.
How do you think: can we trust y-full dates?
I understand, if for some branch they have wrong levels of SNP equivalents of course they will get wrong dates after calculation.
But in case if their levels of SNP equivalents are correct, do they produce more or less correct dates or they have some systematic error?

hartaisarlag
02-02-2020, 07:03 PM
Been away for some time (busy), but I did notice the update and I found it both encouraging and discouraging. I (and others familiar with ZS241) have known for several years that YFull's tree was severely flawed both in terms of dating and leveling of SNP equivalents. The recent update is a big step in the right direction, with Y3088 as a ZS222 equivalent (something I suspected for some time), in effect placing Z18271's TMRCA in the Iron Age and its parent in an LBA context. Another prediction I want to make is that ZS227 is also going to be a Y3088 equivalent, and that the age of the branch under which it is now subsumed (S20075) will also come down.

Now regarding the line of reasoning you put forward, I can understand why a Jewish origin would seem problematic at first, but ultimately that's the most plausible explanation... And not necessarily because the area from which the Brazilian individual hails is known to be a crypto-Jewish reservoir.

There is little doubt in my mind that ZS222 (Y3088) is going to be an essentially Israelite lineage, one might even call it "Proto-Israelite". As I said above, its TMRCA falls in an LBA-IA I context, which is the time frame in which the Israelites emerged.

YFull has dropped its TMRCA estimate for J-Y3088/ZS222 to 1000 BCE, and for J-Z18271 to 800 BCE.

Worth noting that a Crimean Karaite has just joined the Iraqi Jew under J-ZS2458.

A bunch of other interesting phylogenetic developments, on cross-referencing YFull and FTDNA.

StillWater
02-02-2020, 07:32 PM
YFull has dropped its TMRCA estimate for J-Y3088/ZS222 to 1000 BCE, and for J-Z18271 to 800 BCE.

Worth noting that a Crimean Karaite has just joined the Iraqi Jew under J-ZS2458.

A bunch of other interesting phylogenetic developments, on cross-referencing YFull and FTDNA.

Does the Crimean Karaite identify as a Cohen?

Agamemnon
02-02-2020, 07:41 PM
How do you think: can we trust y-full dates?
I understand, if for some branch they have wrong levels of SNP equivalents of course they will get wrong dates after calculation.
But in case if their levels of SNP equivalents are correct, do they produce more or less correct dates or they have some systematic error?

Yes and no. Yes in the sense that there no longer is that 25% inflation you used to see a couple of years ago. No in the sense that it ultimately depends on having a sufficient(ly representative) amount of samples on the tree.

So for example we just had a new update today, Y3088's TMRCA went down from 3400 ybp to 3000 ybp while Z18271's TMRCA went from 2900 ybp to 2800 ybp. As a rule of thumb, ZS222/Y3088's TMRCA should find itself in an LBA-IA I time frame while Z18271 should be within an IA II-C time frame (circa ~2600 yBP) so the last update is a step in the right direction at least for Z18271.

Another similar example would be L862/Z2331's TMRCA which went down from 6400 ybp to 6000 ybp in the latest update. Even more important in my view is the fact that the 6000 ybp TMRCA estimate is shared with 9 other branches under L862/Z2331 while the 6400 ybp estimate was shared with only 3 other branches (major ones though, which is relevant). This is important because it highlights the demographic "explosion" for this lineage which we know occurred during the Late Chalcolithic, L862 and the other major branches immediately downstream should have the TMRCAs close to 5900 yBP, an event I'd attribute to the break-up of Proto-Semitic and the beginning of early Semitic dispersals with nomadic pastoralists from the southernmost parts of the Levant. Likewise Z1853 and Z1865's TMRCA estimates provide an interesting window between 5200-4300 BCE where it is very tempting to place the initial arrival of J lineages and Iran_ChL-type admixture (earliest wave would've been J1-P58 and pre-J2b1-M205) in the Levant, FYI that time frame overlaps with the Pottery Neolithic which (again in my view) is the period where we're likely to see the dissemination of that type of ancestry.

So in this sense, the last update is something of a leap forward and the tree in its current version is far more accurate than the former version. But since it will be recalibrated with new results, this might be a temporary situation and we might end up with a situation where the tree seldom reflects what the mass of the data that is already available seems to be showing.


YFull has dropped its TMRCA estimate for J-Y3088/ZS222 to 1000 BCE, and for J-Z18271 to 800 BCE.

Worth noting that a Crimean Karaite has just joined the Iraqi Jew under J-ZS2458.

A bunch of other interesting phylogenetic developments, on cross-referencing YFull and FTDNA.

Also notice how the Syrian individual is, together with a new Mexican individual from Nuevo Leon, under a new branch: BY55414.
I know for a fact this branch comprises Kohanim from Morocco, Greece and Iraq.

hartaisarlag
02-02-2020, 08:53 PM
Also notice how the Syrian individual is, together with a new Mexican individual from Nuevo Leon, under a new branch: BY55414.
I know for a fact this branch comprises Kohanim from Morocco, Greece and Iraq.

Yeah, not sure the identity of the other new upload; waiting to hear back.

FTDNA is now saying that this branch and FGC17491 (which seems to include Spaniards, Portuguese, and Mexicans, Moroccan Jews, and Ashkenazim) form a clade together: J-FT157560.

FTDNA also reports J-FT34605 as a new branch of J-Z18271, including someone with roots in Turkey (Jewish, I'm assuming) and a Puerto Rican.

hartaisarlag
02-02-2020, 08:59 PM
Another similar example would be L862/Z2331's TMRCA which went down from 6400 ybp to 6000 ybp in the latest update. Even more important in my view is the fact that the 6000 ybp TMRCA estimate is shared with 9 other branches under L862/Z2331 while the 6400 ybp estimate was shared with only 3 other branches (major ones though, which is relevant). This is important because it highlights the demographic "explosion" for this lineage which we know occurred during the Late Chalcolithic, L862 and the other major branches immediately downstream should have the TMRCAs close to 5900 yBP, an event I'd attribute to the break-up of Proto-Semitic and the beginning of early Semitic dispersals with nomadic pastoralists from the southernmost parts of the Levant. Likewise Z1853 and Z1865's TMRCA estimates provide an interesting window between 5200-4300 BCE where it is very tempting to place the initial arrival of J lineages and Iran_ChL-type admixture (earliest wave would've been J1-P58 and pre-J2b1-M205) in the Levant, FYI that time frame overlaps with the Pottery Neolithic which (again in my view) is the period where we're likely to see the dissemination of that type of ancestry.

Somewhat OT, but what do you make of the diversification of E-M84 between 7800 ybp and 6500 ybp? Too early to be tied to PS, and yet, heftier (and better-matched geographically) than E-L791's breakup ca. 5800 ybp.