PDA

View Full Version : Population genomics of Bronze Age Eurasia (Allentoft et al. 2015)



Pages : [1] 2 3 4 5

Jean M
06-10-2015, 06:26 PM
This paper is of huge importance, so I think it deserves its own thread, though it has already been reported on the news thread.

Morten E. Allentoft et al., Population genomics of Bronze Age Eurasia, Nature, 522, 167–172 (11 June 2015)
http://www.nature.com/nature/journal/v522/n7555/full/nature14507.html

Abstract

The Bronze Age of Eurasia (around 3000–1000 BC) was a period of major cultural changes. However, there is debate about whether these changes resulted from the circulation of ideas or from human migrations, potentially also facilitating the spread of languages and certain phenotypic traits. We investigated this by using new, improved methods to sequence low-coverage genomes from 101 ancient humans from across Eurasia. We show that the Bronze Age was a highly dynamic period involving large-scale population migrations and replacements, responsible for shaping major parts of present-day demographic structure in both Europe and Asia. Our findings are consistent with the hypothesized spread of Indo-European languages during the Early Bronze Age. We also demonstrate that light skin pigmentation in Europeans was already present at high frequency in the Bronze Age, but not lactose tolerance, indicating a more recent onset of positive selection on lactose tolerance than previously thought.

4760


By analysing our genomic data in relation to previously published ancient and modern data (Supplementary Information, section 6), we find evidence for a genetically structured Europe during the Bronze Age (Fig. 2; Extended Data Fig. 1; and Supplementary Figs 5 and 6). Populations in northern and central Europe were composed of a mixture of the earlier hunter-gatherer and Neolithic farmer groups, but received ‘Caucasian’ genetic input at the onset of the Bronze Age (Fig. 2). This coincides with the archaeologically well defined expansion of the Yamnaya culture from the Pontic-Caspian steppe into Europe (Figs 1 and 2). This admixture event resulted in the formation of peoples of the Corded Ware and related cultures, as supported by negative ‘admixture’ f3 statistics when using Yamnaya as a source population (Extended Data Table 2, Supplementary Table 12). Although European Late Neolithic and Bronze Age cultures such as Corded Ware, Bell Beakers, Unetice, and the Scandinavian cultures are genetically very similar to each other (Fig. 2), they still display a cline of genetic affinity with Yamnaya, with highest levels in Corded Ware, lowest in Hungary, and central European Bell Beakers being intermediate (Fig. 2b and Extended Data Table 1).


The close affinity we observe between peoples of Corded Ware and Sintashta cultures (Extended Data Fig. 2a) suggests similar genetic
sources of the two, which contrasts with previous hypotheses placing the origin of Sintastha in Asia or the Middle East. Although we
cannot formally test whether the Sintashta derives directly from an eastward migration of Corded Ware peoples or if they share common
ancestry with an earlier steppe population, the presence of European Neolithic farmer ancestry in both the Corded Ware and the Sintashta,
combined with the absence of Neolithic farmer ancestry in the earlier Yamnaya, would suggest the former being more probable (Fig. 2b and
Extended Data Table 1).

Humanist
06-10-2015, 06:30 PM
The following was posted in the other thread my MfA:

http://abload.de/img/nature14507-sf61rwr50.jpg

Mher
06-10-2015, 06:36 PM
Velika Gruda X1 from Montenegro

MfA
06-10-2015, 06:39 PM
irRus: J2
irArm: J2b
irAltai: J2a, Q1a
baAfGo: R1a
baKarasuk: Q1a, R1a
baMezh: R1a, R1b
baAndrov: R1a
baSintashta: R1a
baStalingradQuar: R1b
baYam: I2a, R1b
baArm: E1b, R1b
baSca: I1, I1a, R1a, R1b
baCW: R1, R1a, R1b
baBb: R1b
baHu: G2a, I2, I2a
baRem: I2, I2a

angscoire
06-10-2015, 06:40 PM
The Swedish Viby sample is R1a apparently , but a low resolution result , so no subclade.

R.Rocca
06-10-2015, 06:40 PM
German Bell Beaker sample SAMEA3325421 is P312+, but also Z196+, meaning he is also DF27. What is surprising however is that he is also Z195-. I don't think we have any modern Z196+Z195- samples, but I could be wrong. This of course could be huge for R1b!

ADW_1981
06-10-2015, 06:41 PM
The Stalingrad/Volgograd sample is R1b, and the R1b Armenian samples are from Nerkin Getashen.

parasar
06-10-2015, 06:41 PM
The following was posted in the other thread my MfA:

http://abload.de/img/nature14507-sf61rwr50.jpg

Looks like J2 folk made their move at the cusp of the Iron age - irRus, irArm, irAltai, BR2 (1200bc).

Radboud
06-10-2015, 06:43 PM
irRus: J2
irArm: J2b
irAltai: J2, Q1a
baAfGo: R1a
baKarasuk: Q1a, R1a
baMezh: R1a, R1b
baAndrov: R1a
baSintashta: R1a
baStalingradQuar: R1b
baYam: I2a, R1b
baArm: E1b, R1b
baSca: I1, I1a, R1a, R1b
baCW: R1, R1a, R1b
baBb: R1b
baHu: G2a, I2, I2a
baRem: I2, I2a

What is Mezh?

MfA
06-10-2015, 06:49 PM
What is Mezh?

Mezhovskaya Culture (1300-800/700 BC).

ADW_1981
06-10-2015, 06:52 PM
What is Mezh?

Location is in Bashkortotstan.

Passa
06-10-2015, 07:06 PM
My God 30% E1b1b in Bronze Age Armenians! Six times more than modern Armenians! How I wish they had tested Bronze Age Balkanites too...

Mher
06-10-2015, 07:07 PM
Age of sample?

Mher
06-10-2015, 07:11 PM
no no only one sapmle have E1

MfA
06-10-2015, 07:12 PM
My God 30% E1b1b in Bronze Age Armenians! Six times more than modern Armenians! How I wish they had tested Bronze Age Balkanites too...

Great day for E1b folks for sure. Keep in mind E1b-M84 is around 20% on neighbouring modern Assyrian sub-group of Chaldeans and tad more than 10% on overall Kurds and close to 20% on Dersimi Kurds.

Passa
06-10-2015, 07:15 PM
no no only one sapmle have E1

Still, 30%

alan
06-10-2015, 07:16 PM
R1b in corded ware for first time. Are the Yamnaya R1b new samples?

Humanist
06-10-2015, 07:20 PM
R1b Armenian samples are from Nerkin Getashen.

If I am not mistaken, crania from Nerkin Getashen were studied in the following paper:

On the Origin of the Armenians (In the Light of Non-Metric Cranial Traits Data)
Alla Movsesian and Nvard Kochar
Iran & the Caucasus
Vol. 8, No. 2 (2004), pp. 183-197



We can now postulate the genetic integrity of the contemporary and ancient populations of Armenia, starting from the Bronze period at least. This is corroborated by the data of craniometry, differentiating the contemporary and ancient groups only by value of the cephalic index.


Their (ancient specimens) dolichocrany in this case does not impede the establishment of genetic links through the late development of brachycephalization. The genetic ties between the epochs become even more evident when we examine the data on the discrete varying traits, subject to neither the epochal variation nor to the influence of the environmental factors.


The genetics appears to support what is stated above (assuming the autosomal data is not significantly inconsistent). Also, perhaps another nail in the coffin for the cephalic index as a reliable measure of genetic relatedness through time.

vettor
06-10-2015, 07:22 PM
Looks like Remedello for Mtdna is only ...............H2 , J1 and X2

the only group to have X2

jesus
06-10-2015, 07:22 PM
From West Asia, Iranians and Turks seem to be the closest to the Armenian samples, other than Armenians themselves I'd assume.

http://www.nature.com/nature/journal/v522/n7555/images/nature14507-sf4.jpg

alan
06-10-2015, 07:23 PM
R1a missing from Yamnaya again. Seems the idea that CW and its eastwards chain being linked to R1a flourishing looks correct. Basically tagging R1a to the Anthony concept of the chain of CW-Fatyanovo-Abashevo-Sintashta which has long seemed likely given the modern linguistic-genetic associations.

alan
06-10-2015, 07:25 PM
If I am not mistaken, crania from Nerkin Getashen were studied in the following paper:

On the Origin of the Armenians (In the Light of Non-Metric Cranial Traits Data)
Alla Movsesian and Nvard Kochar
Iran & the Caucasus
Vol. 8, No. 2 (2004), pp. 183-197




Perhaps another nail in the coffin for the cephalic index as a reliable measure of genetic relatedness through time.

That coffin has more nails than wood now

Krefter
06-10-2015, 07:25 PM
LN/BA North Italians were more Neolithic-like(Near Eastern, EEF-shifted) than LN/BA Hungarians. The new Hungarians may have more ANE than the ones from Gamba 2014, who had just alittle more than Basque. But still it seems the Romedello culture or just North Italians 5,000-4,000 years ago, were very EEF. LN/BA Hungarian and North Italian Y DNA is all I2 and G2a.

http://www.nature.com/nature/journal/v522/n7555/fig_tab/nature14507_ST1.html

With ANE K8 in 4mix Bronze age Hungarians from Gamba 2014 fit as the main ancestors of modern Tuscans. Tuscans fit as Bronze age Hungarian+Cypriot. I'm just think out loud. Possible a population similar to LN/BA Hungarians and North Italians went deep into Italy.

Jean M
06-10-2015, 07:27 PM
Age of sample?

The dates of samples are in Supplementary Table 1 in http://www.nature.com/nature/journal/v522/n7555/extref/nature14507-s1.pdf

Heber
06-10-2015, 07:28 PM
The following was posted in the other thread my MfA:

http://abload.de/img/nature14507-sf61rwr50.jpg

Interesting to note that Bell Beaker and Volga (Stalingrad) are R1b.
This is an area associated with kurgans and stelae.
https://www.pinterest.com/gerardcorcoran/european-bronze-age/

Yamnaya expansion east to Tocharian area would appear to support Victor Mair's Tarim Basin Mummies proposal.
https://www.pinterest.com/gerardcorcoran/tocharian-bronze-age/

The high proportion of Armenian R1b might suggest a path to the Balkans.

alan
06-10-2015, 07:28 PM
LN/BA North Italians were more Neolithic-like(Near Eastern, EEF-shifted) than LN/BA Hungarians. The new Hungarians may have more ANE than the ones from Gamba 2014, who had just alittle more than Basque. But still it seems the Romedello culture or just North Italians 5,000-4,000 years ago, were very EEF. LN/BA Hungarian and North Italian Y DNA is all I2 and G2a.

That is not surprising if it is Remedello I but probably damaging, if it is Remedello 2, to ideas of linking the latter with R1b/steppe infiltration. Both would normally be called copper age though but terminology varies across nations.

vettor
06-10-2015, 07:30 PM
R1a missing from Yamnaya again. Seems the idea that CW and its eastwards chain being linked to R1a flourishing looks correct. Basically tagging R1a to the Anthony concept of the chain of CW-Fatyanovo-Abashevo-Sintashta which has long seemed likely given the modern linguistic-genetic associations.

seems like another smack in the head for R1a group

alan
06-10-2015, 07:31 PM
Its starting to look very

Krefter
06-10-2015, 07:34 PM
That is not surprising if it is Remedello I but probably damaging, if it is Remedello 2, to ideas of linking the latter with R1b/steppe infiltration. Both would normally be called copper age though but terminology varies across nations.

Do you think Remedello were unadmixed EEF? I see on the PCA of Fig 1. they cluster close to Sardinians.

Sangarius
06-10-2015, 07:35 PM
If I am not mistaken, crania from Nerkin Getashen were studied in the following paper:

On the Origin of the Armenians (In the Light of Non-Metric Cranial Traits Data)
Alla Movsesian and Nvard Kochar
Iran & the Caucasus
Vol. 8, No. 2 (2004), pp. 183-197




The genetics appears to support what is stated above (assuming the autosomal data is not significantly inconsistent). Also, perhaps another nail in the coffin for the cephalic index as a reliable measure of genetic relatedness through time.

Well, cephalic index can't be a reliable measure of genetic relatedness through time, simply because of the fact that humans have been uniformly dolicocephalic until fairly recently. Ötzi is the earliest mesocephalic human remain afaik.

Sangarius
06-10-2015, 07:40 PM
Do you think Remedello were unadmixed EEF?

BA Remedello clusters with Central European Neolithic (Stuttgart etc): http://www.nature.com/nature/journal/v522/n7555/fig_tab/nature14507_SF1.html

jesus
06-10-2015, 07:45 PM
Genetic structure of ancient Europe and the Pontic-Caspian steppe.

http://www.nature.com/nature/journal/v522/n7555/images/nature14507-f2.jpg

alan
06-10-2015, 07:49 PM
Do you think Remedello were unadmixed EEF? I see on the PCA of Fig 1. they cluster close to Sardinians.

I have had the impression since I read deeper into it that Remedello I metalwork is more linked to Balkans non-steppe groups. The Remedello I dates are too early to have Yamnaya influences or certainly seems unlikely to me. To be honest there has never been any strong case for steppe migration through material culture for any Remedello phase. Its more the social changes seen in Remedello II that have suggested that possibility than specific items.

Sangarius
06-10-2015, 07:50 PM
Genetic structure of ancient Europe and the Pontic-Caspian steppe.

http://www.nature.com/nature/journal/v522/n7555/images/nature14507-f2.jpg

The change from Bronze Age to modern Armenians is quite interesting. Seems like the BA Armenians had more WHG-like admixture, while modern Armenians have more NE admixture, while the Steppe admixture proportion didn't change much.

J Man
06-10-2015, 07:51 PM
Looks like J2 folk made their move at the cusp of the Iron age - irRus, irArm, irAltai, BR2 (1200bc).

It really is starting to look like for sure.

Humanist
06-10-2015, 07:52 PM
Genetic structure of ancient Europe and the Pontic-Caspian steppe.

Are those the Armenian samples, in the Bronze Age PCA? They appear to be shifted toward the Europeans a bit.

Sangarius
06-10-2015, 07:54 PM
Are those the Armenian samples, in the Bronze Age PCA? They appear to be shifted toward the Europeans a bit.

Yeah, as can be seen in the admixture graph, BA Armenians had more WHG-like admixture, pulling them closer to Europeans compared to modern Armenians.

RCO
06-10-2015, 07:56 PM
"For rs12913832, a major determinant of blue versus brown eyes in humans, our results indicate the presence of blue eyes already in Mesolithic hunter-gatherers as previously described33. We find it at intermediate frequency in Bronze Age Europeans, but it is notably absent from the Pontic-Caspian steppe populations, suggesting a high prevalence of brown eyes in these individuals (Fig. 4). The results for rs4988235, which is associated with lactose tolerance, were surprising. Although tolerance is high in present-day northern Europeans, we find it at most at low frequency in the Bronze Age (10% in Bronze Age Europeans; Fig. 4), indicating a more recent onset of positive selection than previously estimated34"

Krefter
06-10-2015, 08:01 PM
Does anyone have access to the paper?

R.Rocca
06-10-2015, 08:03 PM
I have had the impression since I read deeper into it that Remedello I metalwork is more linked to Balkans non-steppe groups. The Remedello I dates are too early to have Yamnaya influences or certainly seems unlikely to me. To be honest there has never been any strong case for steppe migration through material culture for any Remedello phase. Its more the social changes seen in Remedello II that have suggested that possibility than specific items.

The three Remedello samples go across all three phases (3400-2100 BC), are all I2, and all cluster with European Neolithic Farmers. Looks like they are not Yamnaya derived.

J Man
06-10-2015, 08:06 PM
Does anyone have access to the paper?

I do.

Humanist
06-10-2015, 08:07 PM
I do.

Can you post some quotes regarding the autosomal affinities of the ancient Armenian samples? Please. :)

Jean M
06-10-2015, 08:08 PM
Does anyone have access to the paper?

I do and so does anyone with access to my library.

Radboud
06-10-2015, 08:08 PM
I do.

Are there deep subclades of the R1b, R1a etc?

Arbogan
06-10-2015, 08:09 PM
The change from Bronze Age to modern Armenians is quite interesting. Seems like the BA Armenians had more WHG-like admixture, while modern Armenians have more NE admixture, while the Steppe admixture proportion didn't change much.

Any idea what the dark green means? ANE/Steppe/Caucasus admixture?

Jean M
06-10-2015, 08:09 PM
Can you post some quotes regarding the autosomal affinities of the ancient Armenian samples? Please. :)

I have been posting re the Armenian samples on the thread http://www.anthrogenica.com/showthread.php?2120-quot-DNA-and-the-Origins-of-Peoples-The-Armenians-quot-Lecture

ADW_1981
06-10-2015, 08:12 PM
"For rs12913832, a major determinant of blue versus brown eyes in humans, our results indicate the presence of blue eyes already in Mesolithic hunter-gatherers as previously described33. We find it at intermediate frequency in Bronze Age Europeans, but it is notably absent from the Pontic-Caspian steppe populations, suggesting a high prevalence of brown eyes in these individuals (Fig. 4). The results for rs4988235, which is associated with lactose tolerance, were surprising. Although tolerance is high in present-day northern Europeans, we find it at most at low frequency in the Bronze Age (10% in Bronze Age Europeans; Fig. 4), indicating a more recent onset of positive selection than previously estimated34"

3/10 Yamnaya had LCT, (r1b) and 2/4 Mezh had LCT. (has R1b) Looks like there is a real relationship between the movement of R1b to western Europe, and probably a subset of yet identified subclades of mtDNA. The later R1a rich steppe populations did not frequently carry LCT, although thinking deeper it may have been offset by the introduction of the East Asian populations who were not carriers.

J Man
06-10-2015, 08:15 PM
Can you post some quotes regarding the autosomal affinities of the ancient Armenian samples? Please. :)

There really does not seem to be much said for quotes about the ancient Armenian samples except for their relation to the Yamnaya samples and Indo-European languages. The authors definitely favour the steppe origin hypothesis of IE languages.

jeanL
06-10-2015, 08:19 PM
3/10 Yamnaya had LCT, (r1b) and 2/4 Mezh had LCT. (has R1b) Looks like there is a real relationship between the movement of R1b to western Europe, and probably a subset of yet identified subclades of mtDNA. The later R1a rich steppe populations did not frequently carry LCT, although thinking deeper it may have been offset by the introduction of the East Asian populations who were not carriers.

How would the samples from the Basque periphery burials SJAPL and Longar dated to 3000 BC and 2500 BC interact with it, since they have LTC at ~30% too?

jesus
06-10-2015, 08:20 PM
Yeah, as can be seen in the admixture graph, BA Armenians had more WHG-like admixture, pulling them closer to Europeans compared to modern Armenians.

That could mean that modern Armenians have a minor impact from an Iranian or a Caucasian group, pulling them more to the east or north east.

jesus
06-10-2015, 08:24 PM
Any idea what the dark green means? ANE/Steppe/Caucasus admixture? it's the central Asian teal component I guess.

alan
06-10-2015, 08:33 PM
The three Remedello samples go across all three phases (3400-2100 BC), are all I2, and all cluster with European Neolithic Farmers. Looks like they are not Yamnaya derived.

Looking that way. Otzi had links to the Remedello I culture but was a G guy but same basic picture. This also fits of course with the non-R nature of all samples in or around the early copper age of the southern Europe.

alan
06-10-2015, 08:56 PM
So basically pre-2000BC we have R1b confirmed or likely on the L23 derived lines from Yamnaya, beaker and corded ware and its looking unlikely the pre-beaker Med. copper route was used judging by the non-R nature of all the pre-beaker Med. ancient DNA so far. Its looking more and more likely that a route through central Europe was used. It is a small sample though so nothing is certain. I am still thinking, especially if Remedello is eliminated that the only visible culture connecting the the east and west of Europe c. 2750BC is corded ware. I just cannot see any actual culture other than CW bridging the vast gap. If beaker arose c. 2750BC corded ware at that time essentially occupied most of central and northern Europe from Ukraine to the eastern border of France. Of course if extremely low archaeological visibility is an issue then anything is possible.

However, my take is beaker was formed c. 2750-2700BC when a sub-group with close parallels with CW somehow crossed from central Europe into Iberia leaving little trace in France. There is one possible explanation for lack of impact in France and a clue IMO as to the nature of the journey might be in the return journey of beaker eastwards which seems to have headed along the French coast and up the Rhone by 2600BC. It doesnt necessarily follow that this was a reverse revisiting of the ancestral route but it does seem a possibility that a central European group came down the Rhone to the French Med coast and started a network linking to Iberia albeit that some of the formation of beaker culture per-se happened in Iberia before spreading back east. Remember the very high variance of P312 in the Rhone/west Alps sort of area of France too. Other routes are of course possible but I wouldnt know where to start. Bypassing the Pyrenees to the south or north using the coast seems like a good idea compared to the Pyrenees then the whole width of the Iberian intererior - albeit there are handy east to west flowing rivers.

J Man
06-10-2015, 09:05 PM
It would be nice to know how many male samples from Iron Age Russia and Iron Age Altai were tested.

jesus
06-10-2015, 09:08 PM
That could mean that modern Armenians have a minor impact from an Iranian or a Caucasian group, pulling them more to the east or north east.

Modern Armenians are Indeed slightly Iranian or Georgian Jewish shifted

http://www.nature.com/nature/journal/v513/n7518/images_article/nature13673-f2.jpg

VS


http://www.nature.com/nature/journal/v522/n7555/images/nature14507-f2.jpg

George
06-10-2015, 09:15 PM
The three Remedello samples go across all three phases (3400-2100 BC), are all I2, and all cluster with European Neolithic Farmers. Looks like they are not Yamnaya derived.

Any views on the Hungarian I2 and particularly the Yamna I2a? Especially the latter.

ADW_1981
06-10-2015, 09:16 PM
It would be nice to know how many male samples from Iron Age Russia and Iron Age Altai were tested.

I believe there are 5 males. Three are not assigned dates. However the other two are between 396BC-209BC, and 721-889AD. With a little more digging, possibly in the raw data or the ID#s someone can probably figure out which are the J2s and which are the Q1a.

ADW_1981
06-10-2015, 09:18 PM
Any views on the Hungarian I2 and particularly the Yamna I2a? Especially the latter.

Probably a contact zone with the west whom we know had I2 populated in the region. I'd be curious to see what SNPs were positive on these remains

J Man
06-10-2015, 09:20 PM
I believe there are 5 males. Three are not assigned dates. However the other two are between 396BC-209BC, and 721-889AD. With a little more digging, possibly in the raw data or the ID#s someone can probably figure out which are the J2s and which are the Q1a.

Okay thanks. It would have helped a lot to if they could have mentioned which cultures these Iron Age Russian and Iron Age Altai samples came from as well. It would be informative. 721-889 AD looks later than the Iron Age.

George
06-10-2015, 09:23 PM
Probably a contact zone with the west whom we know had I2 populated in the region. I'd be curious to see what SNPs were positive on these remains

Would I be correct in assuming that the Yamna I2a is from the area between Don and Volga? (No problem with the Hungarian location).

Humanist
06-10-2015, 09:23 PM
If the data is available, has anyone determined which varieties of R1b the two Middle Bronze Age Armenians are?

Coldmountains
06-10-2015, 09:34 PM
Wow Sintashta has less of the teal component than Yamnaya and more of the "European" this is quite unexpected but really interesting. Sintashta and Andronovo have R1a like expected but did they wrote to which subclade they belonged?.

Arbogan
06-10-2015, 09:40 PM
Wow Sintashta has less of the teal component than Yamnaya and more of the "European" this is quite unexpected but really interesting. Sintashta and Andronovo have R1a like expected but did they wrote to which subclade they belonged?.

I suspect that their TEAL/south-central-Asian must have been diluted over time by repeated immigrations east-ward from some EHG like population. They seem to be inbetween yamnaya and corded-ware/bell beaker culture admixture wise, with some minor infiltration by Siberian/ENA like admixture.

ADW_1981
06-10-2015, 09:57 PM
I suspect that their TEAL/south-central-Asian must have been diluted over time by repeated immigrations east-ward from some EHG like population. They seem to be inbetween yamnaya and corded-ware/bell beaker culture admixture wise, with some minor infiltration by Siberian/ENA like admixture.

That's a possibility, but it could also be from a lack of interaction with whatever populations were rich in the Teal component. In your scenario, it would require eastward immigration of a very EHG/WHG like population with very low Teal at 2000 BC. This scenario is not as likely in the more recent time frame especially since it would have been in Yamnaya territory. Areas further north and east most likely had less Teal but more EHG, and hadn't *yet* felt the onslaught of the Asiatics ;)

Mher
06-10-2015, 10:03 PM
Armenians R1b 100% come from Yamna

Coldmountains
06-10-2015, 10:05 PM
That's a possibility, but it could also be from a lack of interaction with whatever populations were rich in the Teal component.

The teal component in Yamnaya looks Caucasus derived so it makes sense that Sintashta was lower in it but this also means that Proto-Indo-Iranian ancestry among modern Indo-Iranians is not so high like many assume. I wonder which modern population is closest to Sintashta. It is also possible that Proto-Indo-Iranians mixed much with urban BMAC folks before they entered Afghanistan, India and Iran so that their EHG-like ancestry was already diluted in an early stage but they were quite successful in spreading their Y-DNA in Central and South Asia.

Jean M
06-10-2015, 10:05 PM
http://www.nytimes.com/2015/06/16/science/dna-deciphers-roots-of-modern-europeans.html

The New York Times welcomes the two papers in the current issue of Nature, one of which (Haak et al.) has already been much discussed on this forum, as it was available in pre-print.


DNA Deciphers Roots of Modern Europeans

For centuries, archaeologists have reconstructed the early history of Europe by digging up ancient settlements and examining the items that their inhabitants left behind. More recently, researchers have been scrutinizing something even more revealing than pots, chariots and swords: DNA.

On Wednesday in the journal Nature, two teams of scientists — one based at the University of Copenhagen and one based at Harvard University — presented the largest studies to date of ancient European DNA, extracted from 170 skeletons found in countries from Spain to Russia. Both studies indicate that today’s Europeans descend from three groups who moved into Europe at different stages of history.

The first were hunter-gatherers who arrived some 45,000 years ago in Europe. Then came farmers who arrived from the Near East about 8,000 years ago. Finally, a group of nomadic sheepherders from western Russia called the Yamnaya arrived about 4,500 years ago. The authors of the new studies also suggest that the Yamnaya language may have given rise to many of the languages spoken in Europe today.

Ron Pinhasi, an archaeologist at University College Dublin who was not involved in either study, said that the new studies were “a major game-changer. To me, it marks a new phase in ancient DNA research.”...

David W. Anthony, an archaeologist at Hartwick College and a co-author on the Harvard study, said it was likely that the expansion of Yamnaya into Europe was relatively peaceful. “It wasn’t Attila the Hun coming in and killing everybody,” he said. Instead, Dr. Anthony thought the most likely scenario was that the Yamnaya “entered into some kind of stable opposition” with the resident Europeans that lasted for a few centuries. But then gradually the barriers between the cultures eroded.

alan
06-10-2015, 10:21 PM
The Yamnaya-like nature of Afansievo and the lack of R1a in Yamnaya (but well represented in CW and its derivatives) really makes me think that Afansievo had a big R1b component. This of course raises question marks about the later R1a Tarim people on the more southerly trade route and whether they really related to either Afanasievo or Tocharian or were they part of a later trickle along the southern trade route rather than direct descendants of the Afansievo people who arrived at Altai, in all probability by a slightly more northern steppe route.

There really does seem to be a very strong chain for IE linked R1a leading back to corded ware and it seems to be via that chain of CW-Fatyanovo-Abashevo-Sintashta-Andronovo that R1a spread north and east and became strong linked to Baltic, Slavic and Indo-Iranian to modern times. However this is not true of Afanasievo so I have a strong feeling when they get actual Afansievo yDNA from Altai etc rather than from the Tarim trade route we will see a lot of R1b.

Agamemnon
06-10-2015, 10:55 PM
Still no R1a in Yamnaya (though I2a has been found), this is starting to get very interesting... Also, J2's history looks quite complicated (J2b in Iron Age Armenia and J2 + J2a in the Iron Age samples from Russia and the Altai), to say the least... Though the Iron Age Armenian sample carrying J2b certainly has something to do with the arrival of Proto-Armenian speakers from the west (the Balkans in particular). The high E1b frequencies in the Bronze Age Armenians is totally unexpected, I'm quite eager to discover which subclade(s?) these samples belong to (the same goes for R1b though, given that these R1b samples predate the arrival of Proto-Armenian speakers). Considering how the ancient Armenians plot, I'd say that my theory according to which Cypriots are similar to the pre-exilic Judeans is bound to make some sense.

Arbogan
06-10-2015, 11:14 PM
The teal component in Yamnaya looks Caucasus derived so it makes sense that Sintashta was lower in it but this also means that Proto-Indo-Iranian ancestry among modern Indo-Iranians is not so high like many assume. I wonder which modern population is closest to Sintashta. It is also possible that Proto-Indo-Iranians mixed much with urban BMAC folks before they entered Afghanistan, India and Iran so that their EHG-like ancestry was already diluted in an early stage but they were quite successful in spreading their Y-DNA in Central and South Asia.

Well I think the yamnaya scores gave us a hint. The averae west-Iranian got between 16-19% yamnaya derived, and 12-14% yamnaya. when the teal was excluded. We are about as Indo-European as turks or greeks, more or less.

alan
06-10-2015, 11:15 PM
what date and culture was the BA Scandinavia R1b?

Webb
06-10-2015, 11:16 PM
German Bell Beaker sample SAMEA3325421 is P312+, but also Z196+, meaning he is also DF27. What is surprising however is that he is also Z195-. I don't think we have any modern Z196+Z195- samples, but I could be wrong. This of course could be huge for R1b!

You are correct that there are no modern samples who are Z196+/Z195-. At least in the DF27 project and the 1000 Genome samples. But this is huge news!!!

ADW_1981
06-10-2015, 11:23 PM
Still no R1a in Yamnaya (though I2a has been found), this is starting to get very interesting... Also, J2's history looks quite complicated (J2b in Iron Age Armenia and J2 + J2a in the Iron Age samples from Russia and the Altai), to say the least... Though the Iron Age Armenian sample carrying J2b certainly has something to do with the arrival of Proto-Armenian speakers from the west (the Balkans in particular). The high E1b frequencies in the Bronze Age Armenians is totally unexpected, I'm quite eager to discover which subclade(s?) these samples belong to (the same goes for R1b though, given that these R1b samples predate the arrival of Proto-Armenian speakers). Considering how the ancient Armenians plot, I'd say that my theory according to which Cypriots are similar to the pre-exilic Judeans is bound to make some sense.

Unless I'm reading it inccorectly, the "Iron" age J2b Armenian from Norabak is still about 3000 years old and only a couple hundred years later than the R1b ones. The higher EHG seems linked with an arrival of R1b from the north east. Initially it might be tempting to link J2b to the Balkans but I suspect it's more likely related to a common Mesopotamian origin.

Agamemnon
06-10-2015, 11:33 PM
Unless I'm reading it inccorectly, the "Iron" age J2b Armenian from Norabak is still over 3000 years old and only a couple hundred years later than the R1b ones. The higher EHG seems linked with an arrival of R1b from the north east.

The arrival of Proto-Armenian speakers coincides with the Late Bronze Age collapse and proceeds further during the Iron Age roughly until the 7th century BCE, so even if this sample is ~3000 years old (although I think it's 2700 years old) it still fits within the appropriate time frame.
Also, all of the Bronze Age samples from Armenia were R1b and E1b (M123 from the looks of it, which is rather surprising).

Coldmountains
06-10-2015, 11:37 PM
Well I think the yamnaya scores gave us a hint. The averae west-Iranian got between 16-19% yamnaya derived, and 12-14% yamnaya. when the teal was excluded. We are about as Indo-European as turks or greeks, more or less.

Seems to be true for most Indo-Iranians except of Pamiri Tajiks who indeed have some significant "European/EHG" ancestry. The high frequencies of R1a among Pashtuns and some Indian castes/populations look like founder effects and Proto-Indo-Iranian ancestry peaks today in the Pamir and to a lesser extent in the Hindukush. But curiously some Jatts have much "European" ancestry compared to other Indo-Iranians also.

lgmayka
06-10-2015, 11:43 PM
Still no R1a in Yamnaya (though I2a has been found), this is starting to get very interesting.
This gives rise to the question: Where was R1a before CW? Was it north of Yamnaya?

Arbogan
06-10-2015, 11:47 PM
Seems to be true for most Indo-Iranians except of Pamiri Tajiks who indeed have some significant "European/EHG" ancestry. The high frequencies of R1a among Pashtuns and some Indian castes/populations look like founder effects and Proto-Indo-Iranian ancestry peaks today in the Pamir and to a lesser extent in the Hindukush. But curiously some Jatts have much "European" ancestry compared to other Indo-Iranians also.

I think using the sintastha and androvono will give us an idea.

Agamemnon
06-10-2015, 11:49 PM
This gives rise to the question: Where was R1a before CW? Was it north of Yamnaya?

This is what things are starting to look like, but bear in mind that all of these samples come from the eastern end of the Yamna horizon so far, so perhaps R1a will eventually show up in the more westerly areas of the Yamna horizon... Who knows? Remains from the Pontic-Caspian steppe cultures which predate the Yamna horizon (such as Sredny Stog) would enable us to provide a realistic answer to this question.

Generalissimo
06-10-2015, 11:50 PM
This gives rise to the question: Where was R1a before CW? Was it north of Yamnaya?

In Yamnaya just west and north of the eastern Yamnaya sampled to date?

Another option is that Repin was PIE, and it contributed ancestry to both Corded Ware and Yamnaya proper.

The genomes from the Tarim Basin will be interesting. And Y-DNA from Afansievo.

Arbogan
06-10-2015, 11:54 PM
This is what things are starting to look like, but bear in mind that all of these samples come from the eastern end of the Yamna horizon so far, so perhaps R1a will eventually show up in the more westerly areas of the Yamna horizon... Who knows?

the new yamnayas aren't from kazakistan/Russian border area. They're from the black-sea-pontic-Transcaucasia region.

Generalissimo
06-10-2015, 11:55 PM
the new yamanayas aren't from kazakistan/Russian border area. They're from the black-sea-pontic-Transcaucasia region.

No, they're from the Don region.

Transcaucasia is south of the Caucasus.

Agamemnon
06-10-2015, 11:57 PM
the new yamanayas aren't from kazakistan/Russian border area. They're from the black-sea-pontic-Transcaucasia region.

That still qualifies as one of the easternmost areas of the Yamna horizon (even though Yamnaya never actually covered Transcaucasia, as Generalissimo said the samples are from the Don region).

Piquerobi
06-10-2015, 11:59 PM
Very interesting! Afanasievo is significantly related to Yamnaya (the rumours of R1b being present at Afanasievo now make even more sense!). And Remedello, long thought of as IE derived, is quite Neolithic European (with 0 R1a and 0 R1b). Corded Ware R1b also has been found!

ADW_1981
06-11-2015, 12:06 AM
Taking a quick gander, E-M123 is the most common branch of E in Armenia as per the FTDNA Armenian project, and if I recall tends to be the variation in Iran as well (from recollection on studies). I bet it dates back to this same time period or earlier when the northern Middle East was in contact with the Levant.

Arbogan
06-11-2015, 12:06 AM
That still qualifies as one of the easternmost areas of the Yamna horizon.
It's almost dash in the middle of it. Depending on which parameters and borders accepted. Either way it's interesting that it follows the same pattern as the yamnayas from samara region(which would be considered far eastern).




No, they're from the Don region.

Transcaucasia is south of the Caucasus. .

My bad, I was thinking of transitional region between Caucasus and the pontic steppe.

rms2
06-11-2015, 12:11 AM
The Yamnaya-like nature of Afansievo and the lack of R1a in Yamnaya (but well represented in CW and its derivatives) really makes me think that Afansievo had a big R1b component. This of course raises question marks about the later R1a Tarim people on the more southerly trade route and whether they really related to either Afanasievo or Tocharian or were they part of a later trickle along the southern trade route rather than direct descendants of the Afansievo people who arrived at Altai, in all probability by a slightly more northern steppe route.

There really does seem to be a very strong chain for IE linked R1a leading back to corded ware and it seems to be via that chain of CW-Fatyanovo-Abashevo-Sintashta-Andronovo that R1a spread north and east and became strong linked to Baltic, Slavic and Indo-Iranian to modern times. However this is not true of Afanasievo so I have a strong feeling when they get actual Afansievo yDNA from Altai etc rather than from the Tarim trade route we will see a lot of R1b.

Which brings us back to the mysterious rumored findings of Alexei Kovalev, i.e., R1b in the Altai in Afanasievo and Okunevo remains. Sounds like the rumor could be true, but where's the Russian paper?

everest59
06-11-2015, 12:12 AM
Here is one Sintashta sample results:
ANE K7 ( I don't have the K8)
30.99% ANE
8.25% ASE
43.05% WHG-UHG
0.05% East_Eurasian
1.84% West_African
0.00% East_African
15.84% ENF

everest59
06-11-2015, 12:14 AM
With Eurogenes K7, I managed to get ~30K snp's in common with the calculator.

Passa
06-11-2015, 12:14 AM
Taking a quick gander, E-M123 is the most common branch of E in Armenia as per the FTDNA Armenian project, and if I recall tends to be the variation in Iran as well (from recollection on studies). I bet it dates back to this same time period or earlier when the northern Middle East was in contact with the Levant.

M123 is very very ancient and originated in the Levant. It populated all of West Asia before the arrival of IEs and J-P209 IMO.

alan
06-11-2015, 12:15 AM
As it stands the R1a -CW/CW derived cultures link looks incredibly solid. However its worth noting thaat CW proper has had a mix of R1a, R1b and non-R in the various ancient DNA samples. So I am sort of thinking a founder effect in favour of R1a might have happened early in the CW-Fatyanovo-Abashevo-Sintashta cultural chain and this effect was passed on. Nevetheless so far as the sample grows the link between CW and R1a seems much stronger than Yamnaya which still has no R1a. My feeling since the ancient DNA data came out, and prior to that based on archaeological considerations is that R1a was strong on the Middle Dnieper and and adjacent and lay inland and upstream from the more arid southern and eastern Euro steppes where I suspect R1b predominated. From the directions and environments that Corded Ware spread into, it seems to me that they had an environmental and probably subsistence preference/adaptation for the forest steppe and forest steppe and forest areas. Note that in Swiss studies they found Corded Ware had a pattern of forest living too which contrasted with the bell beaker preference for open pasture.

Arbogan
06-11-2015, 12:15 AM
Here is one Sintashta sample results:
ANE K7 ( I don't have the K8)
30.99% ANE
8.25% ASE
43.05% WHG-UHG
0.05% East_Eurasian
1.84% West_African
0.00% East_African
15.84% ENF

So they're North east-Europeans slightly modified with higher ANE and ASE like admixture. Now we just need some iron age Indo-Iranian sample.

Agamemnon
06-11-2015, 12:17 AM
M123 is very very ancient and originated in the Levant. It populated all of West Asia before the arrival of IEs and J-P209 IMO.

Its main subclade, M84, happens to be the strongest contender for the spread of Semitic among E-M35.1's branches. Its distribution is similar to J1-L862's, for instance. But as you said, E-M123 itself is quite ancient, probably 16,000 years old.

Generalissimo
06-11-2015, 12:17 AM
It's almost dash in the middle of it. Depending on which parameters and borders accepted.

Here's a map of the sampling locations.

Figure 1. (https://drive.google.com/file/d/0B9o3EYTdM8lQenlKRENzR3lpNE0/view?usp=sharing)

Coldmountains
06-11-2015, 12:21 AM
Here is one Sintashta sample results:
ANE K7 ( I don't have the K8)
30.99% ANE
8.25% ASE
43.05% WHG-UHG
0.05% East_Eurasian
1.84% West_African
0.00% East_African
15.84% ENF

Thanks, I am really surprised that they are so high in WHG and so low in ENF. 8% ASE looks quite high but could originate from the" too high" ANE and not show real south Asian affinity. Indo-Iranians changed obviously extremely their genetic structure in the last 4000 years.

R.Rocca
06-11-2015, 12:22 AM
It seems that Corded Ware sample SAMEA3325375 belongs to haplogroup G by way of P187>Z6107+. Not sure why only R1a and R1b show up on Extended Figure 6a. Can anyone else confirm?

I'm trying to give a high level look at the other Corded Ware samples but it looks like their servers are grinding to a halt.

Chad Rohlfsen
06-11-2015, 12:24 AM
Yeah, as can be seen in the admixture graph, BA Armenians had more WHG-like admixture, pulling them closer to Europeans compared to modern Armenians.

Armenians out of the Balkans isn't dead, but looks quite possible now.

alan
06-11-2015, 12:27 AM
This gives rise to the question: Where was R1a before CW? Was it north of Yamnaya?

I think Michal has some thoughts on this. Think he shares my view - once counterintuitive - that R1a had a middle Dnieper hotspot back c. 2900 when CW was forming. I dont think the east-west simple idea for R1a and b difference works now any better than it did before all the Samara R1b. It look much more ecological to me in that R1b may have been more adapted to the arid eastern and southernmost parts of the Euro steppes while R1a was rooted in a more forest steppe economy that could even expand through forested areas.

Agamemnon
06-11-2015, 12:30 AM
Armenians out of the Balkans isn't dead, but looks quite possible now.

From a linguistic standpoint, that's the only model which fits with the data.

alan
06-11-2015, 12:30 AM
It seems that Corded Ware sample SAMEA3325375 belongs to haplogroup G by way of P187>Z6107+. Not sure why only R1a and R1b show up on Extended Figure 6a. Can anyone else confirm?

I'm trying to give a high level look at the other Corded Ware samples but it looks like their servers are grinding to a halt.

there were several non-Rs in Polish CW I recall from a previous ancient DNA paper. Cannot recall if it was I or G.

Arbogan
06-11-2015, 12:30 AM
Thanks, I am really surprised that they are so high in WHG and so low in ENF. 8% ASE looks quite high but could originate from the" too high" ANE and not show real south Asian affinity. Indo-Iranians changed obviously extremely their genetic structure in the last 4000 years.

I'm not really sure they represent indo-Iranians, other than subset of them, that are extremely north bound. I'd be more interested in samples from the central, and southern regions, from later eras.

everest59
06-11-2015, 12:33 AM
I'm not really sure they represent indo-Iranians, other than subset of them, that are extremely north bound. I'd be more interested in samples from the central, and southern regions, from later eras.

I will do Andronovo next. Need to download the BAM file and call variants on it, which will take some time.

Coldmountains
06-11-2015, 12:38 AM
I'm not really sure they represent indo-Iranians, other than subset of them, that are extremely north bound. I'd be more interested in samples from the central, and southern regions, from later eras.

True, the Indo-Iranians who directly contributed to the ancestry of most modern Indo-Iranians were different from Sintashta because of mixture with BMAC and local Central Asians so they were more Near-Eastern and South Central Asian shifted. BMAC was extremely important for the late Proto-Indo-Iranians I think and much of early Iranian and Indian traditions is somewhat derived from them.

Christina
06-11-2015, 12:40 AM
They need more ancient samples from Iberia and modern France. I would venture there will be surprises, like the same clades showing up at the far east and west of Europe.

Megalophias
06-11-2015, 01:08 AM
So given the stubborn lack of Yamnayaness in BA Hungary, despite being full of Kurgans, and in Remedello, is it possible that Western Yamna really was acculturated Cucuteni-Tripolye? Genetically they would then be hardly distinguishable from the Middle Neolithic people they encountered further west.

R.Rocca
06-11-2015, 01:15 AM
So given the stubborn lack of Yamnayaness in BA Hungary, despite being full of Kurgans, and in Remedello, is it possible that Western Yamna really was acculturated Cucuteni-Tripolye? Genetically they would then be hardly distinguishable from the Middle Neolithic people they encountered further west.

BA Hungary looks like it has plenty of the major Yamnaya component.

George
06-11-2015, 01:25 AM
This gives rise to the question: Where was R1a before CW? Was it north of Yamnaya?

Wait until they also check some Yamnas west of the Don. Who knows...

parasar
06-11-2015, 01:39 AM
Anyone knows if Mezhovskaya R1a is Z93+?

Humanist
06-11-2015, 01:45 AM
I will do Andronovo next. Need to download the BAM file and call variants on it, which will take some time.

If you have any spare time, it would be great to see results for one of the MBA Armenians too.

Generalissimo
06-11-2015, 01:53 AM
The Armenians are on the way. It'll be interesting to see how much more steppe ancestry they have than modern Armenians.

parasar
06-11-2015, 01:54 AM
“Now we see the Yamnaya is not only spreading north into Europe; they’re also spreading east, crossing the Urals, getting all the way into central Asia, all the way into the Altai, between Mongolia, China, and Siberia,” says evolutionary biologist Eske Willerslev of the University of Copenhagen...

This eastern branch of the Yamnaya (or Afanasievo) persisted in central Asia and, perhaps, Mongolia and China until they themselves were replaced by fierce warriors in chariots called the Sintashta (also known as the Andronovo culture). These people from the Urals and Caucuses, who were genetically related to central Europeans, persisted in central Asia until 2000 years ago, which means that people in central Asia were actually more like Europeans than living Asians. It wasn’t until relatively recently—just 2000 years ago—that these “Caucasians” were replaced by immigrants from eastern Asia, such as the Karasuk, Mezhovskaya, and other Iron Age cultures that today make up the ancestry of people in central Asia...

“There is a real sense that after more than 2 centuries of linguistics trying to solve the Indo-European question, it's ancient DNA that is suddenly moving us fast toward a possible resolution,” says linguist Paul Heggarty of the Max Planck Institute for Evolutionary Anthropology in Leipzig, Germany. But it is not enough just to have data from northern Eurasia, where the Yamnaya’s movements may reflect only one part of the spread of Indo-European languages. Heggarty adds: “We need key data from the majority of the Indo-European-speaking world in the Mediterranean and south of the Black Sea-Caspian-Himalayas.”

http://news.sciencemag.org/archaeology/2015/06/nomadic-herders-left-strong-genetic-mark-europeans-and-asians

Humanist
06-11-2015, 01:55 AM
The Armenians are on the way.

Nice!! :)

Sein
06-11-2015, 01:55 AM
True, the Indo-Iranians who directly contributed to the ancestry of most modern Indo-Iranians were different from Sintashta because of mixture with BMAC and local Central Asians so they were more Near-Eastern and South Central Asian shifted. BMAC was extremely important for the late Proto-Indo-Iranians I think and much of early Iranian and Indian traditions is somewhat derived from them.

For what it's worth, using formal modelling (based on f-4 stats), modern Indo-Iranians have very substantial levels of Yamnaya-related ancestry.

There is a tendency to look at ADMIXTURE output in these cases, but I think that leads us nowhere. Looking at Chad's ADMIXTURE experiments, that software is very sensitive to slight population/component additions. Drift wrecks havoc with the clusters. At the end of the day, interpreting the output tends to involve a lot of unsubstantiated speculation. Most importantly, there is just no real mechanism for deciding whether the output reflects reality or not.

By contrast, the f4-stat based modelling used in Haak et al. is very robust to things like drift, and doesn't need the "tinkering" involved in ADMIXTURE. Most importantly, the output can be judged on a statistical basis, and is construable as robust, weak, or even as a failure ("infeasible").

I requested David to model the Pashtun and Kalash samples from the HGDP data-set, using the method utilized in Haak et al. (only 10 Pashtun samples, the ones that cluster tightly together, and show some distance to Sindhis and Punjabi Jatts on PCA plots). As always, he was very generous with his time, and was kind enough to try to fit Pashtuns and Kalash as Starcevo_EN + Yamnaya + Dai. The results:

Pashtun
46.7% Yamnaya + 44.1% Starcevo_EN + 9.2% Dai

chisq=0.435

tail probability=0.80468

Kalash
50.1% Yamnaya + 40.1% Starcevo_EN + 9.8% Dai

chisq=0.845

tail probability=0.655534

As you can see, the largest component of Pashtun and Kalash ancestry is Yamnaya. But one also needs to look at the chisq and tail probs. In this case, what we are seeing is exceedingly good. The Pashtun fit has an exceptionally low chisq, and the lower the chisq, the better. It is very hard to get such a good fit. Also, the tail prob is quite high, which is also very good. Interestingly, Pashtuns and Kalash are better modeled as Yamnaya + European Neolithic than most Europeans! This is of considerable importance.

For comparison, David has previously modeled Assyrians as 2% Yamnaya, Armenians as 11% Yamnaya, and Georgians as 22% Yamnaya. Everest was able to model Iranians as 33% Yamnaya. So, the software is showing us a cline that makes a lot of sense. It has Assyrians at only 2%, and then has the Kalash (at the other extreme) at 50%, with Iranians in the middle (but closer to Kalash levels). This is what we would expect, based on linguistics, history, and geography.

Even though I just bashed ADMIXTURE, it is worth noting that Chad's runs consistently had the Yamnaya hijack all the West Eurasian ancestry in peninsular South Asia. They always acted as an "ANI" component for Indians. Surely that gives us some hints, from an ADMIXTURE-based view on things.

Basically, modern Indo-Iranians have very high levels of ancestry from the Indo-European steppe. As is noted in this new paper:

"A genetic link between the Kalash and the steppe through Yamnaya-related people." (Supplementary Information, page 38)

But, it seems that this ancestry didn't come from Sintashta/Andronovo, as those samples look more like the Indo-European ancestors of Europeans than the Indo-European ancestors of West and South/Central Asians. This is all very unexpected, and counter-intuitive. It seems to fly in the face of geography, but we've learned to expect the unexpected whenever it comes to aDNA.

I don't know what to make of the fact that Yamnaya are better proxies for steppe Indo-European ancestry in Asia, rather than Sintashta/Andronovo? Who knows, perhaps the Caucasus route for Indo-Iranian dispersal makes more sense? Also, we don't have autosomal aDNA from people like the Scythians, who might be the direct ancestors of living South Central Asians like Pashtuns. Perhaps they were much more Yamnaya-like, or maybe even closer than Yamnaya are to living South Central Asians? I wouldn't be surprised.

For what it's worth, Afanasievo are almost identical to Yamnaya.

parasar
06-11-2015, 02:01 AM
...

Basically, modern Indo-Iranians have very high levels of ancestry from the Indo-European steppe. As is noted in this new paper:

"A genetic link between the Kalash and the steppe through Yamnaya-related people."

But, it seems that this ancestry didn't come from Sintashta/Andronovo, as those samples look more like the Indo-European ancestors of Europeans than the Indo-European ancestors of West and South/Central Asians. This is all very unexpected, and counter-intuitive. It seems to fly in the face of geography, but we've learned to expect the unexpected whenever it comes to aDNA.

I don't know what to make of the fact that Yamnaya are better proxies for steppe Indo-European ancestry in Asia, rather than Sintashta/Andronovo? Who knows, perhaps the Caucasus route for Indo-Iranian dispersal makes more sense? Also, we don't have autosomal aDNA from people like the Scythians, who might be the direct ancestors of living South Central Asians like Pashtuns. Perhaps they were much more Yamnaya-like, or maybe even closer than Yamnaya are to living South Central Asians? I wouldn't be surprised.

For what it's worth, Afanasievo are almost identical to Yamnaya.

I agree. How about Mezhovskaya?

ADW_1981
06-11-2015, 02:35 AM
Does anyone know if someone is looking at the Y SNP data for these samples? It would be nice to see if deeper analysis can be made that might give better insight on the paper's conclusions. In particular if L51+ can be found on more eastern R1b samples than the usual Bell Beaker.

Krefter
06-11-2015, 02:41 AM
So they're North east-Europeans slightly modified with higher ANE and ASE like admixture. Now we just need some iron age Indo-Iranian sample.

The WHG score is very low, so not like NE Europeans in that way.

Arbogan
06-11-2015, 03:13 AM
For what it's worth, using formal modelling (based on f-4 stats), modern Indo-Iranians have very substantial levels of Yamnaya-related ancestry.

There is a tendency to look at ADMIXTURE output in these cases, but I think that leads us nowhere. Looking at Chad's ADMIXTURE experiments, that software is very sensitive to slight population/component additions. Drift wrecks havoc with the clusters. At the end of the day, interpreting the output tends to involve a lot of unsubstantiated speculation. Most importantly, there is just no real mechanism for deciding whether the output reflects reality or not.

By contrast, the f4-stat based modelling used in Haak et al. is very robust to things like drift, and doesn't need the "tinkering" involved in ADMIXTURE. Most importantly, the output can be judged on a statistical basis, and is construable as robust, weak, or even as a failure ("infeasible").

I requested David to model the Pashtun and Kalash samples from the HGDP data-set, using the method utilized in Haak et al. (only 10 Pashtun samples, the ones that cluster tightly together, and show some distance to Sindhis and Punjabi Jatts on PCA plots). As always, he was very generous with his time, and was kind enough to try to fit Pashtuns and Kalash as Starcevo_EN + Yamnaya + Dai. The results:

Pashtun
46.7% Yamnaya + 44.1% Starcevo_EN + 9.2% Dai

chisq=0.435

tail probability=0.80468

Kalash
50.1% Yamnaya + 40.1% Starcevo_EN + 9.8% Dai

chisq=0.845

tail probability=0.655534

As you can see, the largest component of Pashtun and Kalash ancestry is Yamnaya. But one also needs to look at the chisq and tail probs. In this case, what we are seeing is exceedingly good. The Pashtun fit has an exceptionally low chisq, and the lower the chisq, the better. It is very hard to get such a good fit. Also, the tail prob is quite high, which is also very good. Interestingly, Pashtuns and Kalash are better modeled as Yamnaya + European Neolithic than most Europeans! This is of considerable importance.

For comparison, David has previously modeled Assyrians as 2% Yamnaya, Armenians as 11% Yamnaya, and Georgians as 22% Yamnaya. Everest was able to model Iranians as 33% Yamnaya. So, the software is showing us a cline that makes a lot of sense. It has Assyrians at only 2%, and then has the Kalash (at the other extreme) at 50%, with Iranians in the middle (but closer to Kalash levels). This is what we would expect, based on linguistics, history, and geography.

Even though I just bashed ADMIXTURE, it is worth noting that Chad's runs consistently had the Yamnaya hijack all the West Eurasian ancestry in peninsular South Asia. They always acted as an "ANI" component for Indians. Surely that gives us some hints, from an ADMIXTURE-based view on things.

Basically, modern Indo-Iranians have very high levels of ancestry from the Indo-European steppe. As is noted in this new paper:

"A genetic link between the Kalash and the steppe through Yamnaya-related people." (Supplementary Information, page 38)

But, it seems that this ancestry didn't come from Sintashta/Andronovo, as those samples look more like the Indo-European ancestors of Europeans than the Indo-European ancestors of West and South/Central Asians. This is all very unexpected, and counter-intuitive. It seems to fly in the face of geography, but we've learned to expect the unexpected whenever it comes to aDNA.

I don't know what to make of the fact that Yamnaya are better proxies for steppe Indo-European ancestry in Asia, rather than Sintashta/Andronovo? Who knows, perhaps the Caucasus route for Indo-Iranian dispersal makes more sense? Also, we don't have autosomal aDNA from people like the Scythians, who might be the direct ancestors of living South Central Asians like Pashtuns. Perhaps they were much more Yamnaya-like, or maybe even closer than Yamnaya are to living South Central Asians? I wouldn't be surprised.

For what it's worth, Afanasievo are almost identical to Yamnaya. I think we'll have to be open to the possibility that indo iranian languages diffusion was via linguistic spread rather than genetic diffusion. So far there is nothing clearly pointing to a steppe orgin of modern indo iranian speakers. In the case of western iran it makes sense to me. Everything archaelogically points to pre-iranian orgin, specifically mesopotamian-zagros-elam post neolithic cultures. This is consistent with genetics. Persians and medes and other tribes must have had limited genetic impact. With almost complete absence of EHG. We just need bronze age samples to confirm it. There is no reason for some waves of steppe nomads to replace well developed settlements in the zagros range.

Sein
06-11-2015, 03:20 AM
I think we'll have to be open to the possibility that indo iranian languages diffusion was via linguistic spread rather than genetic diffusion. So far there is nothing clearly pointing to a steppe orgin of modern indo iranian speakers. In the case of western iran it makes sense to me. Everything archaelogically points to pre-iranian orgin, specifically mesopotamian-zagros-elam post neolithic cultures. This is consistent with genetics. Persians and medes and other tribes must have had limited genetic impact. With almost complete absence of EHG. We just need bronze age samples to confirm it.

Certainly, the possibility is there.

But the genetic data argues against this. The best fits for Indo-Iranian West Asians and South Central Asians require Yamnaya, and the amount of Yamnaya-related ancestry is consistent with a history of Indo-Iranian linguistic/cultural identity (I'm referring to Haak et al. modeling, as implemented in qpAdm). By contrast, Semetic West Asians can be modeled as basically 0% Yamnaya. Surely that can't be a coincidence.

We have to ignore ADMIXTURE for a moment, even though that software also shows substantial amounts of Yamnaya-related admixture in South/Central Asia and West Asia.

Edit: Also, for what it's worth, the closest ancient samples to the Kalash, by fst distance, are Karasuk, Yamnaya, and BA Armenians. For Tajiks, Andronovo, Afanansievo, and BA Armenians. That is clearly indicative of something.

Chad Rohlfsen
06-11-2015, 03:23 AM
I think we'll have to be open to the possibility that indo iranian languages diffusion was via linguistic spread rather than genetic diffusion. So far there is nothing clearly pointing to a steppe orgin of modern indo iranian speakers. In the case of western iran it makes sense to me. Everything archaelogically points to pre-iranian orgin, specifically mesopotamian-zagros-elam post neolithic cultures. This is consistent with genetics. Persians and medes and other tribes must have had limited genetic impact. With almost complete absence of EHG. We just need bronze age samples to confirm it.

I'm actually trying a run with Corded serving as a stand-in for Andronovo, until I have the files. Things are looking good. After some tinkering, I may have one to post by the weekend. Btw, ASI is constantly in the 50-60% ballpark for the Paniya.

Arbogan
06-11-2015, 03:34 AM
Certainly, the possibility is there.

But the genetics argues against this. The best fits for Indo-Iranian West Asians and South Central Asians require Yamnaya, and the amount of Yamnaya-related ancestry is consistent with a history of Indo-Iranian linguistic/cultural identity (I'm referring to Haak et al. modeling, as implemented in qpAdm). By contrast, Semetic West Asians can be modeled as basically 0% Yamnaya. Surely that can't be a coincidence.

We have to ignore ADMIXTURE for a moment, even though that software also shows substantial amounts of Yamnaya-related admixture in South/Central Asia and West Asia.

Edit: Also, for what it's worth, the closest ancient samples to the Kalash, by fst distance, are Karasuk and Yamnaya (as well as BA Armenians). There could be several reasons for that. The hypothetical teal like people could have admixed into both. Yamnaya could just be a hypothetical fit that is close enough in resembalance, to a constituent found in south central asians and west asians. Have you tried other fits? Can we say that the right populations have been sampled?

Sein
06-11-2015, 03:37 AM
There could be several reasons for that. The hypothetical teal like people could have admixed into both. Yamnaya could just be a hypothetical fit that is close enough un r resembalance, to constituent found in south central adiand and west asians. Have you tried other fits? Can we say that the right populatiobs have been sampled?

That sort of reticulative population history is a big possibility.

I guess the only option left to us is to wait for some aDNA data from southern Central Asia, Iran, and India.

Arbogan
06-11-2015, 03:46 AM
That sort of reticulative population history is a big possibility.

I guess the only option left to us is to wait for some aDNA data from southern Central Asia, Iran, and India.
Thats the big let down of the entire buisness. While sone groups have it much easier on the virtue of well established academia in paleogenetics. It will take years if it ever happens. And not because of a lack of materials. But because of a general lack of a academic interest. I know that in iran they have some very well preserved specimens. But so far no work has been done on exploring their genetic structure.


http://en.m.wikipedia.org/wiki/Saltmen

parasar
06-11-2015, 03:52 AM
http://dlc.hypotheses.org/807

The title of Haak et al. (2015) writes of the steppe as “a source for Indo-European languages in Europe” [emphasis added], and the abstract claims “These results provide support for the theory of a steppe origin of at least some of the Indo‐European languages of Europe.” But this is not the established Steppe theory for Indo-European, which is instead that the steppe is the origin of all Indo-European languages, and not just of Europe, but everywhere else too: Iran, India, (formerly) Anatolia, and so on. So if the steppe is the source of only some of the Indo-European languages of Europe, then that contradicts the steppe hypothesis.

Allentoft et al. (2015), likewise, make statements such as “Our genomic evidence for the spread of Yamnaya people from the Pontic-Caspian steppe to both northern Europe and Central Asia (Fig. 1) corresponds well with the hypothesized expansion of the Indo-European languages”, when it does not. Northern Europe and Central Asia fit only with a subset of those languages, which is not the Steppe hypothesis. Their title is “Population genomics of Bronze Age Eurasia”, but it too only really addresses northern Eurasia.

Mediterranean Europe, and Asia south of the Black Sea, Caspian and Himalayas, are core areas of Indo-European speech, the large majority of the family, and crucial to answering the Indo-European question. Neither paper does justice to these areas, and neither reports any significant new genetic data for them for the appropriate time-periods.

Ancient DNA and the Indo-European Question
Posted on 2015/06/11 by Paul Heggarty
by Paul Heggarty, Max Planck Institute for Evolutionary Anthropology, Leipzig.

Chad Rohlfsen
06-11-2015, 04:12 AM
There's stuff coming. Be patient.

lgmayka
06-11-2015, 04:13 AM
It seems that Corded Ware sample SAMEA3325375 belongs to haplogroup G by way of P187>Z6107+.
P187 is at the haplogroup F level (http://yfull.com/tree/F/), according to YFull. Z6107 does not exist, according to YFull and YBrowse.

EDIT: The Z-series database shows Z6107 to be 21340310 TA>T (i.e., a deletion), allegedly found in haplogroup G.

Arbogan
06-11-2015, 04:15 AM
There's stuff coming. Be patient. They had north iranian hunther gatherers and some neolithic anatolians. I think those will be generally helpful but outside the scope of the issues regarding indo-iranian speakers.

Williamson
06-11-2015, 04:33 AM
German Bell Beaker sample SAMEA3325421 is P312+, but also Z196+, meaning he is also DF27. What is surprising however is that he is also Z195-. I don't think we have any modern Z196+Z195- samples, but I could be wrong. This of course could be huge for R1b!

I took a look at the BAM file for SAMEA3325421 (RISE563) and it looks to me that he is P312+ U152+, and not Z196+. I haven't identified a further downstream branch as yet. Perhaps someone else should double check this?

Chad Rohlfsen
06-11-2015, 04:50 AM
They had north iranian hunther gatherers and some neolithic anatolians. I think those will be generally helpful but outside the scope of the issues regarding indo-iranian speakers.

That is what we need to separate them from EHG. Then, we can tell what is Yamnaya, and what is local.

Arame
06-11-2015, 05:25 AM
Chad Rohlfsen

The Balkanic theory of Armenians is dead.
In Balkanic theory Armenians came from Balkans after the 1200Bc. And they established quite late in Armenia after the fall of Urartu at 6 BC. During the Bronze Age there were in Anatolia.
So if You assume that those BA Armenians are not Armenian speakers and they are Hurrian speakers this mean that the Hurrians have the affinity with Steppe and not Armenians. But this is impossible because Hurrians are a Neolithic people perhaps even before Kura-Arax culture.
I think everybody will understand now that Armenians/Greeks and some other IE folks directly entered from Yamna to West Asia and started to settle down there pushing neolithic Hurrians to South. Case closed.

Chad Rohlfsen
06-11-2015, 05:28 AM
I'm not going by that date and the European shift looks real.

Arame
06-11-2015, 05:33 AM
Chad
What date You propose? Let's not forget that from 1800-1200 BC Hittite empire was controling all Anatolia and it would be quite dificult to cross Anatolia from Balkans to Armenia.

Arame
06-11-2015, 05:39 AM
Balkanic theory implies that ancient Armenians/Hurrians are more shifted to Near East and later they got some European stuff.
What we see is the contrary on 180 degree. This mean that Balkanic theory is dead and the ancient shift to Europe is just indicating that Armenian R1b came from Europe and not from places like Central Asia.

razyn
06-11-2015, 05:55 AM
I took a look at the BAM file for SAMEA3325421 (RISE563) and it looks to me that he is P312+ U152+, and not Z196+.

He's beginning to sound like a ZZ11* with low coverage sequencing.

Coldmountains
06-11-2015, 06:14 AM
For what it's worth, using formal modelling (based on f-4 stats), modern Indo-Iranians have very substantial levels of Yamnaya-related ancestry.

There is a tendency to look at ADMIXTURE output in these cases, but I think that leads us nowhere. Looking at Chad's ADMIXTURE experiments, that software is very sensitive to slight population/component additions. Drift wrecks havoc with the clusters. At the end of the day, interpreting the output tends to involve a lot of unsubstantiated speculation. Most importantly, there is just no real mechanism for deciding whether the output reflects reality or not.

By contrast, the f4-stat based modelling used in Haak et al. is very robust to things like drift, and doesn't need the "tinkering" involved in ADMIXTURE. Most importantly, the output can be judged on a statistical basis, and is construable as robust, weak, or even as a failure ("infeasible").

I requested David to model the Pashtun and Kalash samples from the HGDP data-set, using the method utilized in Haak et al. (only 10 Pashtun samples, the ones that cluster tightly together, and show some distance to Sindhis and Punjabi Jatts on PCA plots). As always, he was very generous with his time, and was kind enough to try to fit Pashtuns and Kalash as Starcevo_EN + Yamnaya + Dai. The results:

Pashtun
46.7% Yamnaya + 44.1% Starcevo_EN + 9.2% Dai

chisq=0.435

tail probability=0.80468

Kalash
50.1% Yamnaya + 40.1% Starcevo_EN + 9.8% Dai

chisq=0.845

tail probability=0.655534

As you can see, the largest component of Pashtun and Kalash ancestry is Yamnaya. But one also needs to look at the chisq and tail probs. In this case, what we are seeing is exceedingly good. The Pashtun fit has an exceptionally low chisq, and the lower the chisq, the better. It is very hard to get such a good fit. Also, the tail prob is quite high, which is also very good. Interestingly, Pashtuns and Kalash are better modeled as Yamnaya + European Neolithic than most Europeans! This is of considerable importance.

For comparison, David has previously modeled Assyrians as 2% Yamnaya, Armenians as 11% Yamnaya, and Georgians as 22% Yamnaya. Everest was able to model Iranians as 33% Yamnaya. So, the software is showing us a cline that makes a lot of sense. It has Assyrians at only 2%, and then has the Kalash (at the other extreme) at 50%, with Iranians in the middle (but closer to Kalash levels). This is what we would expect, based on linguistics, history, and geography.

Even though I just bashed ADMIXTURE, it is worth noting that Chad's runs consistently had the Yamnaya hijack all the West Eurasian ancestry in peninsular South Asia. They always acted as an "ANI" component for Indians. Surely that gives us some hints, from an ADMIXTURE-based view on things.

Basically, modern Indo-Iranians have very high levels of ancestry from the Indo-European steppe. As is noted in this new paper:

"A genetic link between the Kalash and the steppe through Yamnaya-related people." (Supplementary Information, page 38)

But, it seems that this ancestry didn't come from Sintashta/Andronovo, as those samples look more like the Indo-European ancestors of Europeans than the Indo-European ancestors of West and South/Central Asians. This is all very unexpected, and counter-intuitive. It seems to fly in the face of geography, but we've learned to expect the unexpected whenever it comes to aDNA.

I don't know what to make of the fact that Yamnaya are better proxies for steppe Indo-European ancestry in Asia, rather than Sintashta/Andronovo? Who knows, perhaps the Caucasus route for Indo-Iranian dispersal makes more sense? Also, we don't have autosomal aDNA from people like the Scythians, who might be the direct ancestors of living South Central Asians like Pashtuns. Perhaps they were much more Yamnaya-like, or maybe even closer than Yamnaya are to living South Central Asians? I wouldn't be surprised.

For what it's worth, Afanasievo are almost identical to Yamnaya.

I am also confused because people like Pashtuns, Pamiri and NW Indians have really high levels of R1a so there was a migration of northern steppe people without any doubt. North of Afghanistan is Tajikistan and former BMAC country so we have to expect that Indo-Iranians were not anymore pure when they entered Afghanistan/India and Iran. But at least Tajikistan ist still very close located to the steppe so direct steppe admixture must took there place also. Interestingly Iron Age Russia and Iron Age Altai have J2 what maybe points to a more near eastern/south central asian shift in the Iron Age. The non-east asian part of Kazakh looks like half Tajik/Pashtun and half Russian/Volga Finn so either there was genetic backflow from South Central Asia into the steppe after the Bronze Age or steppe Indo-Iranians of Kazakhstan were somewhat different from Sintashta. We don't know when Pashtuns exactly entered Afghanistan but the later the more different they were from Sintashta I guess. So i agree that having samples of later Scythians or Saka would maybe show which kind of "steppe" ancestry Pashtuns have and I have the feeling that Saka of Central Asia were like northern shifted Pamiri. But we can not exclude that South Central Asia was densely populated or that Indo-Iranian migration was rather small in numbers so that despite of spreading their y-DNA their overall genetic impact was rather small. I am not sure but the high Yamnaya-likeness of Kalash could just originate from the shared "teal" component which was probably already present in South Central Asia prior to Indo-Iranian migration. But maybe I am totally wrong, we have to look now at Andronovo and if possible in the most southern ones maybe they were different from Sintashta and much more Yamnaya/teal-like.

Megalophias
06-11-2015, 06:18 AM
BA Hungary looks like it has plenty of the major Yamnaya component.

Oh, I'm sure it has some. But look at the Y DNA: 3 I2, 1 G2a. No R. More importantly, the D stats: practically every Bronze Age European is closer to Yamna than baHu is. Bell Beaker and Unetice are closer, Corded Ware of course is closer, even Bronze Age Balkans people are closer to Yamna. Conversely, baHu is closer to almost every Bronze Age European than to Yamna, and also to all Neolithic farmers, hunterW, and hunterN. (Also true of Bell Beaker, but to a much smaller degree.)

This is long after the Chalcolithic, so the Yamna ancestry could have been diluted later on. But who did it?

Sein
06-11-2015, 06:59 AM
I am also confused because people like Pashtuns, Pamiri and NW Indians have really high levels of R1a so there was a migration of northern steppe people without any doubt. North of Afghanistan is Tajikistan and former BMAC country so we have to expect that Indo-Iranians were not anymore pure when they entered Afghanistan/India and Iran. But at least Tajikistan ist still very close located to the steppe so direct steppe admixture must took there place also. Interestingly Iron Age Russia and Iron Age Altai have J2 what maybe points to a more near eastern/south central asian shift in the Iron Age. The non-east asian part of Kazakh looks like half Tajik/Pashtun and half Russian/Volga Finn so either there was genetic backflow from South Central Asia into the steppe after the Bronze Age or steppe Indo-Iranians of Kazakhstan were somewhat different from Sintashta. We don't know when Pashtuns exactly entered Afghanistan but the later the more different they were from Sintashta I guess. So i agree that having samples of later Scythians or Saka would maybe show which kind of "steppe" ancestry Pashtuns have and I have the feeling that Saka of Central Asia were like northern shifted Pamiri. But we can not exclude that South Central Asia was densely populated or that Indo-Iranian migration was rather small in numbers so that despite of spreading their y-DNA their overall genetic impact was rather small. I am not sure but the high Yamnaya-likeness of Kalash could just originate from the shared "teal" component which was probably already present in South Central Asia prior to Indo-Iranian migration. But maybe I am totally wrong, we have to look now at Andronovo and if possible in the most southern ones maybe they were different from Sintashta and much more Yamnaya/teal-like.

These are good points.

For what it's worth, after other South Asians, the next closest populations to the "1000 Genomes" Punjabi samples are these (based on fst distances):

Mezhovskaya, LBA (from near the Russo-Kazakh border), fst=0.01

BA, Armenian (who are steppe-shifted in comparison to living Armenians), fst=0.01

Andronovo, fst=0.02

Yamnaya, fst=0.02

So these Punjabis are closer to those ancient samples, rather than to any other living populations outside of South Asia. For what it's worth, these "1000 Genomes" Punjabis are exceedingly South Asian-shifted, and are quite distinct from any of the HGDP Pakistani populations.

The relationship to steppe groups is even evident with the Tamil and Telugu populations! After other South Asians, Tamil and Telugu are closest to Mezhovskaya, Andronovo, and ancient Armenians! I guess this is just something to think about, before we get aDNA from southern Central Asia, Iran, and India.

genetiker
06-11-2015, 07:07 AM
I'm posting the Y haplogroups here:

https://genetiker.wordpress.com/2015/06/11/y-haplogroups-for-prehistoric-eurasian-genomes/

Krefter
06-11-2015, 08:00 AM
I'm posting the Y haplogroups here:

https://genetiker.wordpress.com/2015/06/11/y-haplogroups-for-prehistoric-eurasian-genomes/

Thanks. Here's a sample reference, so you'll know which are male and female and where they are from.

https://docs.google.com/spreadsheets/d/1HuNPykGuq2PbHkUOL5dCiwrveIy-OGO2qOklwfsayW8/edit#gid=540447390

alan
06-11-2015, 08:57 AM
anyone know what part of the BA the Scandinavian samples are from. Curious about R1b there

David Mc
06-11-2015, 09:05 AM
anyone know what part of the BA the Scandinavian samples are from. Curious about R1b there

If I could second that, I am also very curious.

Coldmountains
06-11-2015, 09:09 AM
These are good points.

For what it's worth, after other South Asians, the next closest populations to the "1000 Genomes" Punjabi samples are these (based on fst distances):

Mezhovskaya, LBA (from near the Russo-Kazakh border), fst=0.01

BA, Armenian (who are steppe-shifted in comparison to living Armenians), fst=0.01

Andronovo, fst=0.02

Yamnaya, fst=0.02

So these Punjabis are closer to those ancient samples, rather than to any other living populations outside of South Asia. For what it's worth, these "1000 Genomes" Punjabis are exceedingly South Asian-shifted, and are quite distinct from any of the HGDP Pakistani populations.

The relationship to steppe groups is even evident with the Tamil and Telugu populations! After other South Asians, Tamil and Telugu are closest to Mezhovskaya, Andronovo, and ancient Armenians! I guess this is just something to think about, before we get aDNA from southern Central Asia, Iran, and India.
Very interesting stuff especially for us of Indo-Iranian ancestry at least we have no something from Andronovo/Sintashta and other Central Asian cultures , which is not just Y-DNA and Mtdna. I guess that there will be many surprises in the future. What is about
Mezhovskaya (Date, location, ethnicity?) ? I never heard of them in the past but they have both R1a and R1b and are so much in the east so very interesting folks.

Megalophias
06-11-2015, 09:23 AM
anyone know what part of the BA the Scandinavian samples are from. Curious about R1b there

5/6 are from ~1250-1400 BC (1100-1500 BC including the full confidence intervals), 1/6 is Late Bronze Age (~700 BC).

Jean M
06-11-2015, 09:30 AM
anyone know what part of the BA the Scandinavian samples are from. Curious about R1b there

All the details of the sites of the samples are in the Supplement, which should be open access. If it is not, you have access to it in my library. You also have access to the paper itself.

Jean M
06-11-2015, 09:33 AM
The Centre for GeoGenetics has updated its news page to cover the new paper: http://geogenetics.ku.dk/latest-news/modern-european/


When modern Eurasia was born
BRONZE AGE GENETICS

Modern European and Central Asian peoples are genetically speaking not more than a couple of thousand years old. It was during the Bronze Age that the last major chapters were written in the story of the genetic past of Europe and central Asia. How it happened has been intensively debated among archaeologists. In a study lead by the Centre for GeoGenetics, geneticists and archaeologists from Gothenburg University have generated the largest ancient genomic study to date, and in doing so established how the foundation for modern Europe and Central Asia was laid. Their results are published in the scientific journal Nature ....

Both archaeologists and linguists have had theories about how cultures and languages have spread in our part of the world. We geneticists have now collaborated with them to publish an explanation based on a record amount of DNA-analyses of skeletons from the Bronze Age.

So far the archaeologists have been divided into two different camps. Professor Kristian Kristiansen of the University of Gothenburg, who initiated the project together with Lundbeck Foundation Professor Eske Willerslev says:

- The driving force in our study was to understand the big economical and social changes that happened at the beginning of the third millennium BC, spanning the Urals to Scandinavia. The old Neolithic farming cultures were replaced by a completely new perception of family, property and personhood. I and other archaeologists share the opinion that these changes came about as a result of massive migrations.

With this new investigation the researchers confirm that the changes came about as a result of migrations. The researchers think that this is interesting also because later developments in the Bronze Age are a continuation of this new social perception. Things add up because the migrations can also explain the origin of the northern European language families. ....

Previously the common belief was that lactose tolerance developed in the Balkans or in the Middle East in connection with the introduction of farming during the Stone Age. But now we can see that even late in the Bronze Age the mutation that gives rise to the tolerance is rare in Europe. We think that it may have been introduced into Europe with the Yamnaya herders from Caucasus but that the selection that has made most Europeans lactose tolerant has happened at a much later time.

There is a video of Eske Willersev discussing their findings.

King
06-11-2015, 09:44 AM
The following was posted in the other thread my MfA:

http://abload.de/img/nature14507-sf61rwr50.jpg

What was the sample size for this (for each category)? I apologize in advance if this has already been asked and answered.

Krefter
06-11-2015, 09:48 AM
The most useful part of this paper is probably Supplementary Material Table 10 (http://www.nature.com/nature/journal/v522/n7555/extref/nature14507-s5.xlsx). Allentoft 2015 doesn't present much of their analysis. Besides that we'll have to wait for genome bloggers to analysises.

Krefter
06-11-2015, 09:49 AM
What was the sample size for this (for each category)? I apologize in advance if this has already been asked and answered.

https://drive.google.com/open?id=1HuNPykGuq2PbHkUOL5dCiwrveIy-OGO2qOklwfsayW8&authuser=0

King
06-11-2015, 09:51 AM
Thxs kref.

rms2
06-11-2015, 11:21 AM
Does anyone know if someone is looking at the Y SNP data for these samples? It would be nice to see if deeper analysis can be made that might give better insight on the paper's conclusions. In particular if L51+ can be found on more eastern R1b samples than the usual Bell Beaker.

Yes, that is what I would really like to know, as well.

J Man
06-11-2015, 11:23 AM
https://drive.google.com/open?id=1HuNPykGuq2PbHkUOL5dCiwrveIy-OGO2qOklwfsayW8&authuser=0

Awesome work Kreft. I see that there are 5 Iron Age samples from Russia (including Altai) and that they managed to get Y-DNA from 4 of them. By chance do you know which samples from Iron Age Russia and Iron Age Altai are the J2 and J2a ones?

ArmandoR1b
06-11-2015, 12:11 PM
For those of you wanting to analyze the Y SNP data I assume that you can download the BAM files from http://www.ebi.ac.uk/ena/data/view/PRJEB9021 then use the BAM Analysis Kit on them. You can download the program at http://www.y-str.org/2014/04/bam-analysis-kit.html

rms2
06-11-2015, 12:12 PM
They need more ancient samples from Iberia and modern France. I would venture there will be surprises, like the same clades showing up at the far east and west of Europe.

Anything can happen, I guess, but a lot depends on whether you are talking about y-dna or mtDNA or both, what clades you have in mind, and the ages of the samples.

I could be wrong, but I don't think we are going to see a pattern of general European population continuity, because that would not make sense in light of the tremendous changes that occurred with the onset of the Late Neolithic/Bronze Age period. Among other things, somehow a big shot of ANE was introduced into the population of Europe and it began speaking Indo-European languages.

But I suspect even after every last spadeful of Iberian and French soil is turned, there will still be some who believe that Mesolithic Western European R1b is still out there somewhere waiting to be unearthed.

Mher
06-11-2015, 12:31 PM
Chad Rohlfsen

The Balkanic theory of Armenians is dead.
In Balkanic theory Armenians came from Balkans after the 1200Bc. And they established quite late in Armenia after the fall of Urartu at 6 BC. During the Bronze Age there were in Anatolia.
So if You assume that those BA Armenians are not Armenian speakers and they are Hurrian speakers this mean that the Hurrians have the affinity with Steppe and not Armenians. But this is impossible because Hurrians are a Neolithic people perhaps even before Kura-Arax culture.
I think everybody will understand now that Armenians/Greeks and some other IE folks directly entered from Yamna to West Asia and started to settle down there pushing neolithic Hurrians to South. Case closed.
100% ;)

Chad Rohlfsen
06-11-2015, 01:00 PM
Chad
What date You propose? Let's not forget that from 1800-1200 BC Hittite empire was controling all Anatolia and it would be quite dificult to cross Anatolia from Balkans to Armenia.

They weren't controlling all of Anatolia, and their start is closer to 1600BCE. Plus, we don't know who these people are in Bronze Age Armenia. They could be any other group that enters into the Levant and Mesopotamia. R1b heavy areas are outside of Hittite territory, at least today. Hittite also needn't be the only IE language being spoken in the empire, but that of the elites. There is a lot of gray area here. L23 isn't just heavy in Armenians, but several people, and higher diversity on the west coast of the Black Sea. Plus, Armenians have several younger European R1b clades, likely from the Balkans. BTW, crossing the sea is not difficult. Neolithic farmers were doing it 5,000 years prior to these samples. We have kurgan-like burials in Greek Islands after 2800BCE. So, you can't say it's dead.

Agamemnon
06-11-2015, 01:02 PM
Chad Rohlfsen

The Balkanic theory of Armenians is dead.
In Balkanic theory Armenians came from Balkans after the 1200Bc. And they established quite late in Armenia after the fall of Urartu at 6 BC. During the Bronze Age there were in Anatolia.
So if You assume that those BA Armenians are not Armenian speakers and they are Hurrian speakers this mean that the Hurrians have the affinity with Steppe and not Armenians. But this is impossible because Hurrians are a Neolithic people perhaps even before Kura-Arax culture.
I think everybody will understand now that Armenians/Greeks and some other IE folks directly entered from Yamna to West Asia and started to settle down there pushing neolithic Hurrians to South. Case closed.

The Balkanic model for the arrival of Armenian is anything but dead, as I said it's the only one which fits with the linguistic data. Again, this is a linguistic issue, no amount of ancient DNA is going to change this (much like the PIE urheimat debate for that matter). In light of these results, it seems that R1b came with Anatolian speakers instead of Proto-Armenian speakers, and that they were subsequently assimilated by the neighbouring Hurrians.
Also, I fail to see how Hurrians are "neolithic".

J Man
06-11-2015, 01:09 PM
M123 is very very ancient and originated in the Levant. It populated all of West Asia before the arrival of IEs and J-P209 IMO.

Where do you think J-P209 came from?

smal
06-11-2015, 01:23 PM
RISE397_Kapan_LBA_Armenia L23+ > Y4371/Z8128+ > Z2106+ CTS9219-
RISE555_Stalingrad Quarry_EBA_Russia PF6399/S10+ > CTS7340/Z2107+ > Z2106+

Mher
06-11-2015, 01:23 PM
They weren't controlling all of Anatolia, and their start is closer to 1600BCE. Plus, we don't know who these people are in Bronze Age Armenia. They could be any other group that enters into the Levant and Mesopotamia. R1b heavy areas are outside of Hittite territory, at least today. Hittite also needn't be the only IE language being spoken in the empire, but that of the elites. There is a lot of gray area here. L23 isn't just heavy in Armenians, but several people, and higher diversity on the west coast of the Black Sea. Plus, Armenians have several younger European R1b clades, likely from the Balkans. BTW, crossing the sea is not difficult. Neolithic farmers were doing it 5,000 years prior to these samples. We have kurgan-like burials in Greek Islands after 2800BCE. So, you can't say it's dead.
Armenian several R1b1 from galatians .

Passa
06-11-2015, 01:27 PM
Where do you think J-P209 came from?

Iran or around it.

Mher
06-11-2015, 01:28 PM
RISE397_Kapan_LBA_Armenia L23+ > Y4371/Z8128+ > Z2106+ CTS9219-
RISE555_Stalingrad Quarry_EBA_Russia PF6399/S10+ > CTS7340/Z2107+ > Z2106+
Kapan?from Kapan was only two women

Dr_McNinja
06-11-2015, 01:32 PM
Here is one Sintashta sample results:
ANE K7 ( I don't have the K8)
30.99% ANE
8.25% ASE
43.05% WHG-UHG
0.05% East_Eurasian
1.84% West_African
0.00% East_African
15.84% ENF

For comparison,

Yamnaya in K7:
34.76 ANE, 4.45 ASE, 8.17 ENF, 52.27 WHG, 0.69 West African (from Gedmatch of Yamnaya kits)

Karelia_HG in K7:
37.52 ANE, 3.63 ASE, 55.15 WHG, 3.70 East Eurasian

Sintashta looks like the beginning of the bridge to South Central Asia. It could be modeled as half Yamnaya and half a population that was 27.22 ANE, 12.05 ASE, 33.83 WHG, 23.51 ENF, 2.99 West African. That's like one admixture event away from some modern South Asians. Half this non-Yamnaya-half-of-Sintashta and half Gedrosian component (35 ANE, 9.28 ASE, 55.11 ENF) results in "31.315 ANE, 10.665 ASE, 16.915 WHG, 39.31 ENF, 1.5 West African" which is almost at modern Haryana Jatt (about 3-4% swing in WHG and ASE).

Krefter
06-11-2015, 01:33 PM
RISE397_Kapan_LBA_Armenia L23+ > Y4371/Z8128+ > Z2106+ CTS9219-
RISE555_Stalingrad Quarry_EBA_Russia PF6399/S10+ > CTS7340/Z2107+ > Z2106+

How did you get their Y SNP calls? Can you do the Andronovo and Sintashta R1a next?

Krefter
06-11-2015, 01:35 PM
For comparison,

Yamnaya in K7:
34.76 ANE, 4.45 ASE, 8.17 ENF, 52.27 WHG, 0.69 West African (from Gedmatch of Yamnaya kits)

Karelia_HG in K7:
37.52 ANE, 3.63 ASE, 55.15 WHG, 3.70 East Eurasian

Sintashta looks like the beginning of the bridge to South Central Asia. It could be modeled as half Yamnaya and half a population that was 27.22 ANE, 12.05 ASE, 33.83 WHG, 23.51 ENF, 2.99 West African. That's like one admixture event away from some modern South Asians. Half this non-Yamnaya-half-of-Sintashta and half Gedrosian component (35 ANE, 9.28 ASE, 55.11 ENF) results in "31.315 ANE, 10.665 ASE, 16.915 WHG, 39.31 ENF, 1.5 West African" which is almost at modern Haryana Jatt (about 3-4% swing in WHG and ASE).

If that Sintashta's ANE K7 score has nothing to do with a low amount of SNPs, and is confirmed to have relation to South Asians that Yamnaya lacks, where would you think it comes from?

smal
06-11-2015, 01:35 PM
Kapan?from Kapan was only two women
I'm not sure about sex, but Y chromosome is available.

Mher
06-11-2015, 01:39 PM
I'm not sure about sex, but Y chromosome is available.

Oh yesssssssssss WOW https://docs.google.com/spreadsheets/d/1HuNPykGuq2PbHkUOL5dCiwrveIy-OGO2qOklwfsayW8/edit?pli=1#gid=1729003196

Dr_McNinja
06-11-2015, 01:43 PM
If that Sintashta's ANE K7 score has nothing to do with a low amount of SNPs, and is confirmed to have relation to South Asians that Yamnaya lacks, where would you think it comes from?

We'd need ancient South Central Asian DNA to be sure. BMAC and something from Afghanistan as well.

More likely there were already descendants of South Central Asians with higher ASE living there when these European-like people arrived from the western steppe (or wherever... the study says there was a movement from North Europe straight to the Caspian-Aral area steppe, skipping Pontic steppe and Yamnaya?).

Coldmountains
06-11-2015, 01:49 PM
For comparison,

Yamnaya in K7:
34.76 ANE, 4.45 ASE, 8.17 ENF, 52.27 WHG, 0.69 West African (from Gedmatch of Yamnaya kits)

Karelia_HG in K7:
37.52 ANE, 3.63 ASE, 55.15 WHG, 3.70 East Eurasian

Sintashta looks like the beginning of the bridge to South Central Asia. It could be modeled as half Yamnaya and half a population that was 27.22 ANE, 12.05 ASE, 33.83 WHG, 23.51 ENF, 2.99 West African. That's like one admixture event away from some modern South Asians. Half this non-Yamnaya-half-of-Sintashta and half Gedrosian component (35 ANE, 9.28 ASE, 55.11 ENF) results in "31.315 ANE, 10.665 ASE, 16.915 WHG, 39.31 ENF, 1.5 West African" which is almost at modern Haryana Jatt (about 3-4% swing in WHG and ASE).
Ok thats interesting. I as amateur was confused by the low ENF admixture among them but it is almost twice as much as in Yamnaya and they are lower in WHG. But if i am not wrong they have less of the teal-like component so some of their ENF admixture is EEF-like( CT ?) or i am wrong?

J Man
06-11-2015, 01:52 PM
Iran or around it.

I think so to. Possibly the Zagros area or a bit further east.

Agamemnon
06-11-2015, 01:57 PM
I think so to. Possibly the Zagros area or a bit further east.

I think it's too early to say for sure.

J Man
06-11-2015, 01:57 PM
I think it's too early to say for sure.

It is...Just my guess.

Coldmountains
06-11-2015, 02:00 PM
We'd need ancient South Central Asian DNA to be sure. BMAC and something from Afghanistan as well.

More likely there were already descendants of South Central Asians with higher ASE living there when these European-like people arrived from the western steppe (or wherever... the study says there was a movement from North Europe straight to the Caspian-Aral area steppe, skipping Pontic steppe and Yamnaya?).
Can you model Sintashta as mixture of modern populations? even when the results will be not really accurate maybe we get so a clue what kind of mixture of ancient populations they were

Anabasis
06-11-2015, 02:08 PM
Arent there any admixture results fof Ancients samples from Armenia? Any of calculator exsist in Gedmatch is oke.

alan
06-11-2015, 02:46 PM
I'm not sure about sex, but Y chromosome is available.

Its overrated. Just have a cup of tea :0)

everest59
06-11-2015, 03:01 PM
Whoops. I accidentally posted the RISE505 Andronovo results in the ancient DNA news thread. Here are the results:

ANE K7
30.01% ANE
8.04% ASE
37.56% WHG-UHG
5.03% East_Eurasian
0.00% West_African
0.00% East_African
19.36% ENF


Harappaworld:
10.36% S-Indian
25.85% Baloch
5.21% Caucasian
42.78% NE-Euro
0.22% SE-Asian
1.88% Siberian
0.01% NE-Asian
0.01% Papuan
1.30% American
3.35% Beringian
7.51% Mediterranean
0.00% SW-Asian
0.00% San
0.00% E-African
0.24% Pygmy
1.26% W-African

everest59
06-11-2015, 03:28 PM
Alright, I just did Afanasievo RISE507. Harappaworld was only able to use 7k snp's:
12.20% S-Indian
43.94% Baloch
2.23% Caucasian
35.37% NE-Euro
0.00% SE-Asian
0.00% Siberian
0.00% NE-Asian
0.00% Papuan
3.36% American
2.90% Beringian
0.00% Mediterranean
0.00% SW-Asian
0.00% San
0.00% E-African
0.00% Pygmy
0.00% W-African

ANE K7
48.33% ANE
10.31% ASE
27.49% WHG-UHG
0.00% East_Eurasian
0.00% West_African
0.00% East_African
13.87% ENF

Probably very noisy.

nuadha
06-11-2015, 03:34 PM
more like the south central part of the yamnaya horizon. To draw so close to these new samples in order to group them with the other eastern samples is coaxing the data. We should not presume a strict east west divide. These new samples just covered a lot of new territory and pretty much central in the yamnaya horizon.

Joe B
06-11-2015, 03:41 PM
RISE397_Kapan_LBA_Armenia L23+ > Y4371/Z8128+ > Z2106+ CTS9219-
RISE555_Stalingrad Quarry_EBA_Russia PF6399/S10+ > CTS7340/Z2107+ > Z2106+Z2106+ for RISE397_Kapan_LBA_Armenia is interesting. Wonder if he is CTS7763+ or Z2109+? Then again he could be negative for both like K2015-GS000018396-Tabas17 (Tabasarans, Dagestan_Makhachkala) is.

Any SNP results downstream of Z2106 for RISE555_Stalingrad Quarry_EBA_Russia?


I'm not sure about sex, but Y chromosome is available.Lola - The Kinks (https://www.youtube.com/watch?v=LemG0cvc4oU)

George
06-11-2015, 03:51 PM
irRus: J2
irArm: J2b
irAltai: J2a, Q1a
baAfGo: R1a
baKarasuk: Q1a, R1a
baMezh: R1a, R1b
baAndrov: R1a
baSintashta: R1a
baStalingradQuar: R1b
baYam: I2a, R1b
baArm: E1b, R1b
baSca: I1, I1a, R1a, R1b
baCW: R1, R1a, R1b
baBb: R1b
baHu: G2a, I2, I2a
baRem: I2, I2a

Unless I've missed something, there is no table included specifying which result pertains to which sample (unlike the mtDNA Supplemental table 14). Right?

Chad Rohlfsen
06-11-2015, 04:13 PM
This South Eurasian is skewing away from the truth. ASI is lacking in this region. I think it has to do more with something old within Central Asia that may carry over Crown Eurasian stuff which makes them more related to Native Americans and South Asians. For instance, MA1 is pretty much intermediary between EHG and Ust-Ishim. Looking at PCA's the direction and distance of drift from Kostenki to Loschbour, and MA1 to Karelia, is fairly similar. If we had a European that was 12ky closer to Loschbour, it may be the same. It's as if the separation of East and West Eurasians was a constant change that continued up until the time of our samples, and not so much about admixture from one, lacking in the other, but a step in a long process. This border region of West and East Eurasians may have a good amount of Ust-Ishim type stuff or mimics it quite well. How much is really East or West Eurasian, or neither, will take a few samples from Kazakhstan down through Tajikistan. I may try to separate Ust-Ishim from my ASI cluster, just to check it.

Dubhthach
06-11-2015, 04:26 PM
Don't know if this has been posted but link to this counter argument by Paul Heggerty (Farmers vs. Steepe) was tweeted to me (In Irish no less ;) )
https://twitter.com/KingDonncha1/status/609024078775631872?t=1&cn=bWVudGlvbg%3D%3D&sig=5d8a97636b669c87b65725d80ff0d7926a6d2c64&al=1&refsrc=email&iid=e14b5015967d4909b48f5edd7855c7dd&autoactions=1434037630&uid=21091698&nid=4+1489


http://dlc.hypotheses.org/807

parasar
06-11-2015, 04:41 PM
Here is one Sintashta sample results:
ANE K7 ( I don't have the K8)
30.99% ANE
8.25% ASE
43.05% WHG-UHG
0.05% East_Eurasian
1.84% West_African
0.00% East_African
15.84% ENF

"Andronovo
Eurogenes ANE K7:
30.01% ANE
8.04% ASE
37.56% WHG-UHG
5.03% East_Eurasian
0.00% West_African
0.00% East_African
19.36% ENF"

Are you seeing any pattern? Andronovo and Sintashta look very similar.

Afanasievo as expected looks more Baikal oriented. Is there any reason to believe that Afanasievo has a Yamna origin? To me it looks it could also be the other way around.

"Afanasievo
ANE K7
48.33% ANE
10.31% ASE
27.49% WHG-UHG
0.00% East_Eurasian
0.00% West_African
0.00% East_African
13.87% ENF"

everest59
06-11-2015, 04:45 PM
I don't trust the Afasanievo results for that sample fully, as we are dealing with very low SNP count. So, I will be calling variants on RISE511, which is a 10 gig file. It should give a much better picture.
But, yeah, I think one would expect Andronovo and Sintashta to be similar.
However, unless it's noise, Andronovo seems to have some East Asian admixture. The Andronovo data I used is a pretty good one.

ADW_1981
06-11-2015, 04:49 PM
The paper draws the conclusion that Andronovo may have been descended from Europe based on a small sliver of Neolithic Farmer ancestry found in both. I have not gone back to the Admixture graphs to see this, but it must have been very small because I couldn't make it out, or I may not have been looking at the same one. However, wouldn't it be a more logical conclusion that Andronovo had contact with BMAC in the south who already acquired Neolithic Farmer ancestry from Mesopotamian-like people moving east across Iran? This is about 2000 BC after all and long after the spread of farming from the Near East.

The admixture data shows that Sintashta was majority EHG-like in origin without much admixture from the Caucasus or Near East. This only re-affirms the "elite-dominance" model of this component among modern Indo-Iranian speakers. It may have only been a transfer of a R1a1 Y signature that draws a parallel.

everest59
06-11-2015, 04:49 PM
I uploaded the Andronovo sample above to GEDMATCH, perhaps 2 hours ago, but tokenizing is taking too long.

Chad Rohlfsen
06-11-2015, 04:52 PM
I think that I could make a synthetic Afanasievo.

MfA
06-11-2015, 04:56 PM
I think we should make a data sink or something whoever finishes converting a bam file should upload there so that bring some speed to the process.

Helgenes50
06-11-2015, 05:16 PM
I uploaded the Andronovo sample above to GEDMATCH, perhaps 2 hours ago, but tokenizing is taking too long.

Thanks for uploading.
Could you please tell us the kit number ?

Sangarius
06-11-2015, 05:18 PM
I think we should make a data sink or something whoever finishes converting a bam file should upload there so that bring some speed to the process.

Good idea. Who else besides everest59 is converting the bam files right now? I just finished downloading the Armenian bam files, most of them are rather small, the smallest one 168mb, and the largest one around 1.6gb. I also downloaded the BAM analysis kit from Felix and just running the smallest Armenian bam file. Will see how long it takes, it finished chr7 right now..

everest59
06-11-2015, 05:20 PM
Thanks for uploading.
Could you please tell us the kit number ?

Definitely, but let me make sure, it is running right. I can't even use ADMIXTURE right now.
The eye color program they have seems to suggest blue eyes, but it appears the file may not have any eye color alleles. I am not too sure until tokenizing is complete.

everest59
06-11-2015, 05:22 PM
I have the Armenian RISE397 on queue. Not a big file at all. However, the Afanasievo RISE511 sample needs to be complete first. It will take approximately four hours, as this is a huge file.

MfA
06-11-2015, 05:22 PM
Good idea. Who else besides everest59 is converting the bam files right now? I just finished downloading the Armenian bam files, most of them are rather small, the smallest one 168mb, and the largest one around 1.6gb. I also downloaded the BAM analysis kit from Felix and just running the smallest Armenian bam file. Will see how long it takes, it finished chr7 right now..

I'm about to finish RISE416 and 413. also converting 408 but it takes a few more hours.

That 1.6gb file #423 probably going to take a hundred gig or so while processing, so be prepared.

Sein
06-11-2015, 05:24 PM
Whoops. I accidentally posted the RISE505 Andronovo results in the ancient DNA news thread. Here are the results:

ANE K7
30.01% ANE
8.04% ASE
37.56% WHG-UHG
5.03% East_Eurasian
0.00% West_African
0.00% East_African
19.36% ENF


Harappaworld:
10.36% S-Indian
25.85% Baloch
5.21% Caucasian
42.78% NE-Euro
0.22% SE-Asian
1.88% Siberian
0.01% NE-Asian
0.01% Papuan
1.30% American
3.35% Beringian
7.51% Mediterranean
0.00% SW-Asian
0.00% San
0.00% E-African
0.24% Pygmy
1.26% W-African

Okay, my whole bit about Andronovo being more like the Indo-European ancestors of Europeans, rather than South Asians, seems to have been incorrect.

It seems that the closest ancient populations to South Asians are the Andronovo, by fst distances. Also, the HarappaWorld results are much more South Asian-shifted, in comparison to Yamnaya. I think I read too much into the paper's claim that a strong relationship exists between Corded Ware and Andronovo.

smal
06-11-2015, 05:31 PM
Z2106+ for RISE397_Kapan_LBA_Armenia is interesting. Wonder if he is CTS7763+ or Z2109+?

There are no reads for CTS7763, Z2109, or CTS7822. Only few markers are really readable. And I need much more time for an analysis. There was not enough time even to download bam files yet.

ADW_1981
06-11-2015, 05:33 PM
Genetiker has more specific SNPs posted at his blog

https://genetiker.wordpress.com/

R1b definitely has an east to west gradient. If we could reliably determine these as "terminal" SNPs for lack of a better word, it could be quite interesting. Bell Beaker falling into three groups descended from L51.

RISE566 Czech Republic Bell Beaker R1b1a2a1a-P310/PF6546/S129
RISE560 Germany Bell Beaker R1b1a2-L150.1/PF6274.1/S351.1
RISE564 Germany Bell Beaker R1b1a2a1-L51/M412/PF6536/S167

The Russian samples appear to be M269*, and the Armenian one P25*, and M269*.

Would be great to get a second set of eyes, and not sure if these are the terminal SNPs yielded from the samples.
EDIT: As per smal's reference it doesn't look like it.

Sangarius
06-11-2015, 05:33 PM
I have the Armenian RISE397 on queue. Not a big file at all. However, the Afanasievo RISE511 sample needs to be complete first. It will take approximately four hours, as this is a huge file.

Cool. I guess running a 10 gig bam file requires quite a powerful rig?


I'm about to finish RISE416 and 413. also converting 408 but it takes a few more hours.

That 1.6gb file #423 probably going to take a hundred gig or so while processing, so be prepared.

That's great. Thanks for the heads up. The size should be no problem. I wonder about the processing time. I guess I will run Rise423 later on the desktop, it's much more powerful than my laptop I'm currently using.

Coldmountains
06-11-2015, 05:35 PM
Okay, my whole bit about Andronovo being more like the Indo-European ancestors of Europeans, rather than South Asians, seems to have been incorrect.

It seems that the closest ancient populations to South Asians are the Andronovo, by fst distances. Also, the HarappaWorld results are much more South Asian-shifted, in comparison to Yamnaya. I think I read too much into the paper's claim that a strong relationship exists between Corded Ware and Andronovo.

It looks like that they found EEF-admixture among Sintashta but claiming just because of that an Corded Ware origin for them is very speculative in my opinion. From a linguistic point of view Indo-Iranian languages are not closest related to Balto-Slavic languages and closer to Greek. We should also remember that the Andronovo samples were in the very north of the Andronovo horizon so more southern samples will differ from this samples and will show a higher affinity to modern Indo-Iranians

Silesian
06-11-2015, 05:39 PM
Whoops. I accidentally posted the RISE505 Andronovo results in the ancient DNA news thread. Here are the results:
ANE K7
30.01% ANE
8.04% ASE
37.56% WHG-UHG
5.03% East_Eurasian
0.00% West_African
0.00% East_African
19.36% ENF


Harappaworld:
10.36% S-Indian
25.85% Baloch
5.21% Caucasian
42.78% NE-Euro
0.22% SE-Asian
1.88% Siberian
0.01% NE-Asian
0.01% Papuan
1.30% American
3.35% Beringian
7.51% Mediterranean
0.00% SW-Asian
0.00% San
0.00% E-African
0.24% Pygmy
1.26% W-African

Eurogenes-K7-Yamnaya samples

Kit Number: M951285 Iteration: 1000 Delta-Q: 4.795874e-05 Elapsed Time: 21.28 seconds
Population
ANE 35.26%
ASE 4.10%
WHG-UHG 53.76%
East_Eurasian -
West_African 0.31%
East_African -
ENF 6.57%

Kit Number: M020637 Iteration: 1000 Delta-Q: 5.830124e-02 Elapsed Time: 15.98 seconds
Population
ANE 34.26%
ASE 4.80%
WHG-UHG 50.78%
East_Eurasian -
West_African 0.38%
East_African -
ENF 9.77%


HarappaWorld Oracle results:
23 April 2013 - Oracle reference population percentages revised.

Kit M951285

Admix Results (sorted):

# Population Percent
1 NE-Euro 59.99
2 Baloch 29.98
3 Caucasian 4.66
4 American 2.85
5 Beringian 1.88
6 Mediterranean 0.54
7 San 0.05
8 W-African 0.05

HarappaWorld Oracle results:
23 April 2013 - Oracle reference population percentages revised.

Kit M020637

Admix Results (sorted):

# Population Percent
1 NE-Euro 58.9
2 Baloch 29.07
3 Caucasian 6.37
4 Beringian 1.98
5 American 1.85
6 Mediterranean 1.24
7 Papuan 0.38
8 San 0.2

R.Rocca
06-11-2015, 05:49 PM
Genetiker has more specific SNPs posted at his blog

https://genetiker.wordpress.com/

R1b definitely has an east to west gradient. If we could reliably determine these as "terminal" SNPs for lack of a better word, it could be quite interesting. Bell Beaker falling into three groups descended from L51.

RISE566 Czech Republic Bell Beaker R1b1a2a1a-P310/PF6546/S129
RISE560 Germany Bell Beaker R1b1a2-L150.1/PF6274.1/S351.1
RISE564 Germany Bell Beaker R1b1a2a1-L51/M412/PF6536/S167

The Russian samples appear to be M269*, and the Armenian one P25*, and M269*.

Would be great to get a second set of eyes, and not sure if these are the terminal SNPs yielded from the samples.
EDIT: As per smal's reference it doesn't look like it.

I think you took care of this with your edit, but for clarity, these are not terminal as the samples are very low coverage and most of the downstream locations do not have any coverage/data.

parasar
06-11-2015, 05:50 PM
Genetiker has more specific SNPs posted at his blog

https://genetiker.wordpress.com/

R1b definitely has an east to west gradient. If we could reliably determine these as "terminal" SNPs for lack of a better word, it could be quite interesting. Bell Beaker falling into three groups descended from L51.


RISE566 Czech Republic Bell Beaker R1b1a2a1a-P310/PF6546/S129
RISE560 Germany Bell Beaker R1b1a2-L150.1/PF6274.1/S351.1
RISE564 Germany Bell Beaker R1b1a2a1-L51/M412/PF6536/S167

The Russian samples appear to be M269*, and the Armenian one P25*, and M269*.

Would be great to get a second set of eyes, and not sure if these are the terminal SNPs yielded from the samples.



Based on the R1a1 SNPs he has listed as of now, I do not believe they are terminal SNPs.

RISE512 Russia Andronovo R1a1a1b-S224/Z645
RISE566 Czech Republic Bell Beaker R1b1a2a1a-P310/PF6546/S129
RISE560 Germany Bell Beaker R1b1a2-L150.1/PF6274.1/S351.1
RISE564 Germany Bell Beaker R1b1a2a1-L51/M412/PF6536/S167
RISE434 Germany Corded Ware P1
RISE436 Germany Corded Ware CT
RISE446 Germany Corded Ware R1a1a1-M417
RISE1 Poland Corded Ware CT
RISE431 Poland Corded Ware K(xLT)
RISE555 Russia EBA R1b1a2a2-CTS7340/Z2107
RISE492 Russia Iron Age R1
RISE494 Russia Karasuk R1a1a1-Page7
RISE397 Armenia LBA R1b1a2a2-Y4371/Z8128
RISE408 Armenia LBA J
RISE598 Lithuania LBA BT
RISE553 Russia LBA P
RISE554 Russia LBA I?
RISE374 Hungary Maros G2a2a1a-PF6308
RISE413 Armenia MBA R1b1-M415/PF6251
RISE416 Armenia MBA DE
RISE524 Russia Meshovskaya R1b1a2-PF6494
RISE525 Russia Meshovskaya R1a1a1-Page7
RISE21 Denmark Nordic BA ?
RISE47 Denmark Nordic BA R1b1a2-M520/PF6410
RISE175 Sweden Nordic BA I
RISE207 Sweden Nordic BA I1-M450/S109
RISE276 Denmark Nordic LBA R1b1a2-L265/PF6431
RISE42 Denmark Nordic LN CT
RISE179 Sweden Nordic LN I
RISE546 Russia Pit Grave R1b1a2-PF6482/YSC0000203
RISE486 Italy Remedello I2a1a1a-L672/S327
RISE487 Italy Remedello I

J Man
06-11-2015, 05:51 PM
I wonder if the Iron Age Altai J2a and Q1a samples come from early Turkic or proto-Turkic speaking cultures?

can't_lurk_no_mo'
06-11-2015, 05:59 PM
I wonder if the National Genographic Project, FTDNA and 23andMe are going to modify their haplogroup summaries as a result of this latest paper.

Sein
06-11-2015, 06:18 PM
It looks like that they found EEF-admixture among Sintashta but claiming just because of that an Corded Ware origin for them is very speculative in my opinion. From a linguistic point of view Indo-Iranian languages are not closest related to Balto-Slavic languages and closer to Greek. We should also remember that the Andronovo samples were in the very north of the Andronovo horizon so more southern samples will differ from this samples and will show a higher affinity to modern Indo-Iranians

Here are the oracle results for Andronovo:

[1,] "47.2% bhatia_harappa_2 + 52.8% lithuanian_behar_10" "4.9297"
[2,] "55.7% lithuanian_behar_10 + 44.3% sindhi_hgdp_24" "5.0829"
[3,] "51.5% lithuanian_behar_10 + 48.5% punjabi-jatt_harappa_8" "5.5154"
[4,] "55.3% lithuanian_behar_10 + 44.7% punjabi-arain_xing_25" "5.7863"
[5,] "53.4% haryana-jatt_harappa_5 + 46.6% lithuanian_behar_10" "5.7911"
[6,] "42.5% bhatia_harappa_2 + 57.5% russian_hgdp_25" "5.8461"
[7,] "47.9% burusho_hgdp_25 + 52.1% lithuanian_behar_10" "5.8778"
[8,] "60.5% russian_hgdp_25 + 39.5% sindhi_hgdp_24" "5.9062"
[9,] "51.6% lithuanian_behar_10 + 48.4% pathan_hgdp_23" "5.9656"
[10,] "47.4% kalash_hgdp_23 + 52.6% lithuanian_behar_10" "6.2188"

If I recall correctly, this set of populations for Yamnaya would yield 70% NE European + 30% South Asian, while Andronovo appears to be 50/50.

vettor
06-11-2015, 06:33 PM
Genetiker has more specific SNPs posted at his blog

https://genetiker.wordpress.com/

R1b definitely has an east to west gradient. If we could reliably determine these as "terminal" SNPs for lack of a better word, it could be quite interesting. Bell Beaker falling into three groups descended from L51.

RISE566 Czech Republic Bell Beaker R1b1a2a1a-P310/PF6546/S129
RISE560 Germany Bell Beaker R1b1a2-L150.1/PF6274.1/S351.1
RISE564 Germany Bell Beaker R1b1a2a1-L51/M412/PF6536/S167

The Russian samples appear to be M269*, and the Armenian one P25*, and M269*.

Would be great to get a second set of eyes, and not sure if these are the terminal SNPs yielded from the samples.
EDIT: As per smal's reference it doesn't look like it.

thanks, but how accurate is Genetiker?

he has a K( xLT ) for poland sample ( Rise431 ), when no K was noted in the paper that I can find.

K ( xLT ) = the new marker of K-M526 ( or also called the new K2 ) ....which in Karafet paper of 2014 is south-east Asian in origin

Coldmountains
06-11-2015, 06:36 PM
Here are the oracle results for Andronovo:

[1,] "47.2% bhatia_harappa_2 + 52.8% lithuanian_behar_10" "4.9297"
[2,] "55.7% lithuanian_behar_10 + 44.3% sindhi_hgdp_24" "5.0829"
[3,] "51.5% lithuanian_behar_10 + 48.5% punjabi-jatt_harappa_8" "5.5154"
[4,] "55.3% lithuanian_behar_10 + 44.7% punjabi-arain_xing_25" "5.7863"
[5,] "53.4% haryana-jatt_harappa_5 + 46.6% lithuanian_behar_10" "5.7911"
[6,] "42.5% bhatia_harappa_2 + 57.5% russian_hgdp_25" "5.8461"
[7,] "47.9% burusho_hgdp_25 + 52.1% lithuanian_behar_10" "5.8778"
[8,] "60.5% russian_hgdp_25 + 39.5% sindhi_hgdp_24" "5.9062"
[9,] "51.6% lithuanian_behar_10 + 48.4% pathan_hgdp_23" "5.9656"
[10,] "47.4% kalash_hgdp_23 + 52.6% lithuanian_behar_10" "6.2188"

If I recall correctly, this set of populations for Yamnaya would yield 70% NE European + 30% South Asian, while Andronovo appears to be 50/50.
Thanks, the non-east Asian ancestry of Kazakh had similar oracle results (half south Central asian/half eastern European). So the results for Andronovo are in the end not such a big surprise and close to what I expected. They are clearly different from Corded Ware even when both are certainly in some way connected.

ADW_1981
06-11-2015, 06:37 PM
I wonder if the National Genographic Project, FTDNA and 23andMe are going to modify their haplogroup summaries as a result of this latest paper.

Of course not. Didn't you hear? They're partying like it's 2001.

Tolan
06-11-2015, 06:37 PM
Originally Posted by everest59 View Post
Whoops. I accidentally posted the RISE505 Andronovo results in the ancient DNA news thread. Here are the results:

ANE K7
30.01% ANE
8.04% ASE
37.56% WHG-UHG
5.03% East_Eurasian
0.00% West_African
0.00% East_African
19.36% ENF


Harappaworld:
10.36% S-Indian
25.85% Baloch
5.21% Caucasian
42.78% NE-Euro
0.22% SE-Asian
1.88% Siberian
0.01% NE-Asian
0.01% Papuan
1.30% American
3.35% Beringian
7.51% Mediterranean
0.00% SW-Asian
0.00% San
0.00% E-African
0.24% Pygmy
1.26% W-African


Okay, my whole bit about Andronovo being more like the Indo-European ancestors of Europeans, rather than South Asians, seems to have been incorrect.

It seems that the closest ancient populations to South Asians are the Andronovo, by fst distances. Also, the HarappaWorld results are much more South Asian-shifted, in comparison to Yamnaya. I think I read too much into the paper's claim that a strong relationship exists between Corded Ware and Andronovo.

Well, the 7.51 Mediterranean may indicate an European origin, which may be the link with CW!

Megalophias
06-11-2015, 06:39 PM
thanks, but how accurate is Genetiker?

he has a K( xLT ) for poland sample ( Rise431 ), when no K was noted in the paper that I can find.

K ( xLT ) = the new marker of K-M526 ( or also called the new K2 ) ....which in Karafet paper of 2014 is south-east Asian in origin

All R, Q, N, and O belong to K(xLT). Presumably this is R. He is not saying it is K*.

Christina
06-11-2015, 07:33 PM
I am completely agnostic on the big debates in terms of where many of the different HGs originated and spread. My guess is that the picture will be far more complicated than people assume, with periods of substantial drift, and founder effect in certain areas.

I will note though that before we draw any broad conclusions, we need a similar study on the aDNA of Western Europe.

And we need terminal SNPs.

Chad Rohlfsen
06-11-2015, 07:35 PM
Andronovo can have both MN European ancestry and more EHG. There could be something more Amerindian shifted than EHG in South Siberia. Corded did backtrack towards the Samara Valley, Catacomb likely had MN ancestry and Globular Amphora was in the Steppes before 2200 BCE. There's plenty of chances for MN ancestry in Central Asia, on top of something closer to Amerindians than EHG or MA1. I think that is more likely, given the location and just how Afanasievo looks. If Andronovo is more MN like, then it is more Caucasus like and will then be closer to South Asians. For instance, when I use Iranian Jews as a component, it peaks at 80% in the Brahui and Balochi, still over 60% in the Sindhi. This is higher than Yamnaya scores, simply because South Asians are quite a bit more Near Eastern than Yamnaya. Draw a line from Lithuanians to Jatts, and you end up between Yamnaya and the Caucasus. BTW, I did a run and Corded was chosen over Yamnaya by South Asians. I'm throwing in a synthetic Afanasievo too.

ArmandoR1b
06-11-2015, 07:39 PM
thanks, but how accurate is Genetiker?

he has a K( xLT ) for poland sample ( Rise431 ), when no K was noted in the paper that I can find.

K ( xLT ) = the new marker of K-M526 ( or also called the new K2 ) ....which in Karafet paper of 2014 is south-east Asian in origin

Rise431 is baCw which only has R1, R1a, and R1b according to http://www.nature.com/nature/journal/v522/n7555/fig_tab/nature14507_SF6.html

Chad Rohlfsen
06-11-2015, 07:50 PM
Just for reference, Sindhis were about 60% Corded, 20% Bedouin, and 20% ASI. I'm doing the run with an Afanasievo now.

Dr_McNinja
06-11-2015, 07:59 PM
Whoops. I accidentally posted the RISE505 Andronovo results in the ancient DNA news thread. Here are the results:

ANE K7
30.01% ANE
8.04% ASE
37.56% WHG-UHG
5.03% East_Eurasian
0.00% West_African
0.00% East_African
19.36% ENF


Harappaworld:
10.36% S-Indian
25.85% Baloch
5.21% Caucasian
42.78% NE-Euro
0.22% SE-Asian
1.88% Siberian
0.01% NE-Asian
0.01% Papuan
1.30% American
3.35% Beringian
7.51% Mediterranean
0.00% SW-Asian
0.00% San
0.00% E-African
0.24% Pygmy
1.26% W-African

Andronovo's getting even closer to a candidate for a Steppe population that went into South Asia.

It's interesting how similar all these groups are. Mostly HG/ANE derived. Then the peak in ANE in Siberia. I wonder where the East Eurasian in Andronovo is from.

Dr_McNinja
06-11-2015, 08:03 PM
Andronovo can have both MN European ancestry and more EHG. There could be something more Amerindian shifted than EHG in South Siberia. Corded did backtrack towards the Samara Valley, Catacomb likely had MN ancestry and Globular Amphora was in the Steppes before 2200 BCE. There's plenty of chances for MN ancestry in Central Asia, on top of something closer to Amerindians than EHG or MA1. I think that is more likely, given the location and just how Afanasievo looks. If Andronovo is more MN like, then it is more Caucasus like and will then be closer to South Asians. For instance, when I use Iranian Jews as a component, it peaks at 80% in the Brahui and Balochi, still over 60% in the Sindhi. This is higher than Yamnaya scores, simply because South Asians are quite a bit more Near Eastern than Yamnaya. Draw a line from Lithuanians to Jatts, and you end up between Yamnaya and the Caucasus. BTW, I did a run and Corded was chosen over Yamnaya by South Asians. I'm throwing in a synthetic Afanasievo too.Was that in the presence of another high ANE component? From my fooling around it seems like South Asians are drawn to any ANE-heavy component. The higher (going to at least Yamnaya levels) the better. It sometimes even "overlooks" WHG/ENF as long as the ANE is there, though ENF has the greater pull in South Asia obviously.

razyn
06-11-2015, 08:09 PM
German Bell Beaker sample SAMEA3325421 is P312+, but also Z196+, meaning he is also DF27. What is surprising however is that he is also Z195-. I don't think we have any modern Z196+Z195- samples, but I could be wrong. This of course could be huge for R1b!

If true -- I would agree. Have you recanted this (from post #6, and I assume it's too late to edit that, now), in the light of what Alex said in post #126 on this thread? Or is there still some ambivalence, whether it should be called as U152+ or [a subclade of] DF27+? I was also interested in your take on three other Bell Beaker samples, today (in post #197); but there you were commenting on ADW's short list, that didn't happen to include this sample (aka RICE563).

Last night at dinnertime I was all abuzz with excitement, having been the 5th person to test positive for Z196 after FTDNA offered that SNP test in 2011 (when I was a YDNA newbie). By the time I got home from an evening rehearsal, and caught up with this thread, it was already a good bit less certain that there was all that much for DF27 to grasp, here. Yet.

I'm still waiting for the archaeologically inclined among us on this forum to take seriously the iconoclastic quibble (that rms2 recently found and posted) with the received gospel that Atlantic (simple) Bell Beakers (the ceramic objects themselves, not the "archaeological culture" or horizon) are by their very nature earlier than more stylistically developed beakers found farther east -- roughly in the overlap zone with Corded Ware (and almost 50% closer to the steppe). I think that quibble may also get some support from these old bones, and SNPs -- just as others think they are genetically homing in on the migratory routes of PIE speakers into peninsular Europe.

R.Rocca
06-11-2015, 08:15 PM
If true -- I would agree. Have you recanted this...

I will check into it in a few hours. If the script I ran had an issue with Z196 because it was an INDEL, or if it missed the U152 due to a low or heterozygous read.

Chad Rohlfsen
06-11-2015, 08:22 PM
McNinja,

Yes, it was. I had a Yamnaya component too. South Asians don't have as much northern ancestry as MA1 tricked people into believing. It is too old, therefore closer to Ust_Ishim and ASI. Its age also makes it look like it has Basal Eurasian, compared to younger samples. Native Americans are doing the same thing, choosing Yamnaya over MA1. So, MA1 is behaving like something more Near Eastern than Yamnaya. Remember, ANI is basically like Georgians, so it's a ways from Yamnaya. If Andronovo is basically 50% Corded and 50% Afanasievo, that should be easy enough to make. I'll bet that'll make the Sindhi about 40-50% Andronovo, but that will probably change downwards once we have an HG from the area.

Heber
06-11-2015, 08:27 PM
I took a look at the BAM file for SAMEA3325421 (RISE563) and it looks to me that he is P312+ U152+, and not Z196+. I haven't identified a further downstream branch as yet. Perhaps someone else should double check this?


One of the issues is that the phylogenetic tree used to report the branches in these studies is quiet limited compared to those developed by Mike Walsh, Alex Williamson, ISOGG or YFull.
It is by mining the BAM files for significant terminal SNPs we can get a better idea of their position in the tree.
R1b-U152 appears to make sense in this location.
4799

The expansion of R1b appears to have occured below L11 and especially P312. It will be progress to finally get Bronze Age U106, L21, DF27 and U152.
https://www.pinterest.com/gerardcorcoran/european-bronze-age/

vettor
06-11-2015, 08:38 PM
Rise431 is baCw which only has R1, R1a, and R1b according to http://www.nature.com/nature/journal/v522/n7555/fig_tab/nature14507_SF6.html

yes, 60% chance it is R1a and 15% chance each for the others R1 or R1b

Dr_McNinja
06-11-2015, 08:43 PM
McNinja,

Yes, it was. I had a Yamnaya component too. South Asians don't have as much northern ancestry as MA1 tricked people into believing. It is too old, therefore closer to Ust_Ishim and ASI. Its age also makes it look like it has Basal Eurasian, compared to younger samples. Native Americans are doing the same thing, choosing Yamnaya over MA1. So, MA1 is behaving like something more Near Eastern than Yamnaya. Remember, ANI is basically like Georgians, so it's a ways from Yamnaya. If Andronovo is basically 50% Corded and 50% Afanasievo, that should be easy enough to make. I'll bet that'll make the Sindhi about 40-50% Andronovo, but that will probably change downwards once we have an HG from the area.

Here's a few kits I ran through a Eurogenes-ANE-K7 derived calculator (a little noisy at 1d-5):


HRP0349

12.22% ASE
5.74% East_Eurasian
0.11% Yamnaya
0.27% Sintashta
0.55% Andronovo
32.76% Afanasievo
43.97% West-Asian
0.88% Mideast
0.00% Europe
0.49% West-African
2.98% East-African

HRP0393 Haryana Jatt

9.81% ASE
2.77% East_Eurasian
2.33% Yamnaya
3.02% Sintashta
4.18% Andronovo
36.38% Afanasievo
34.66% West-Asian
3.81% Mideast
0.31% Europe
0.08% West-African
2.65% East-African

HRP0341

10.04% ASE
3.12% East_Eurasian
0.14% Yamnaya
0.29% Sintashta
0.61% Andronovo
33.89% Afanasievo
48.12% West-Asian
1.05% Mideast
0.00% Europe
0.00% West-African
2.73% East-African

HRP0370 Afghan Pashtun

5.23% ASE
2.66% East_Eurasian
3.05% Yamnaya
4.05% Sintashta
5.30% Andronovo
21.48% Afanasievo
44.39% West-Asian
11.65% Mideast
0.99% Europe
0.16% West-African
1.05% East-African
West Asian is basically Eurogenes K15 West Asian, around ~26% ANE, rest ENF.

ArmandoR1b
06-11-2015, 09:06 PM
yes, 60% chance it is R1a and 15% chance each for the others R1 or R1b

Right, and looking at Supplementary Table 9 there are only two male CW from Poland and three from Germany. What Genetiker has and what is at http://www.nature.com/nature/journal/v522/n7555/fig_tab/nature14507_SF6.html are very different. Genetiker needs to figure out what he is doing wrong.

Genetiker has -
RISE434 Germany Corded Ware P1
RISE436 Germany Corded Ware CT
RISE446 Germany Corded Ware R1a1a1-M417
RISE1 Poland Corded Ware CT
RISE431 Poland Corded Ware K(xLT)

vettor
06-11-2015, 09:29 PM
Right, and looking at Supplementary Table 9 there are only two male CW from Poland and three from Germany. What Genetiker has and what is at http://www.nature.com/nature/journal/v522/n7555/fig_tab/nature14507_SF6.html are very different. Genetiker needs to figure out what he is doing wrong.

Genetiker has -
RISE434 Germany Corded Ware P1
RISE436 Germany Corded Ware CT
RISE446 Germany Corded Ware R1a1a1-M417
RISE1 Poland Corded Ware CT
RISE431 Poland Corded Ware K(xLT)

What is Genetiker's CT mean

Although the node that joins DE and CF is often referred to as
CT (in the sense that it encompasses the whole of haplogroups C to
T), and the clade that encompasses B, DE, and CF is often referred
to as BT (in the sense that it encompasses the whole of haplogroups
B to T), we are not in favor of these labels as they are inconsistent
with the meanings of other two-letter labels such as CF (referring to
the joint root of C and F, not the range C to F) and LT (referring to
the joint root of L and T, not the range L to T). Instead, we therefore
preferthemoreintuitivelabelsCDEFandBCDEF,respectiv ely,as
used in previous publications [Heinrich et al., 2009; Underhill and
Kivisild, 2007; van Oven et al., 2011].
Within the large K-M9 clade, which encompasses haplogroups
L, M, NO, P (including Q and R), S, and T, a deep bifurcation was
revealed, with haplogroups M, NO, P, and S falling in one subclade,
defined by marker M526 [Chiaroni et al., 2009], and haplogroups L
and T falling in the other subclade, defined by marker P326 [Mendez
et al., 2011]. The latter subclade logically being labeled LT [Mendez
et al., 2011], we “downgraded” the K label to designate the other,
M526-defined subclade, and used KLT for the unifying M9-defined
clade, as proposed previously [van Oven et al., 2011].

http://www.researchgate.net/publication/258117220_Seeing_the_Wood_for_the_Trees_A_Minimal_ Reference_Phylogeny_for_the_Human_Y_Chromosome

The person who creates the Y tree states CT as every haplogroup from C to T

also from 2013,.......... K is purely the M526 branch and K9 is the LT branch ..............that's how I read it.

everest59
06-11-2015, 09:32 PM
Afanasievo RISE511. It does look like the Baloch is higher. This was at >50k snp's:

11.07% S-Indian
32.47% Baloch
3.71% Caucasian
44.54% NE-Euro
0.00% SE-Asian
1.97% Siberian
0.00% NE-Asian
0.00% Papuan
3.57% American
1.81% Beringian
0.00% Mediterranean
0.00% SW-Asian
0.00% San
0.00% E-African
0.04% Pygmy
0.83% W-African

ANE K7 at 70K snp's:

37.88% ANE
11.19% ASE
33.92% WHG-UHG
1.91% East_Eurasian
0.00% West_African
0.00% East_African
15.11% ENF

parasar
06-11-2015, 09:43 PM
Afanasievo RISE511...
ANE K7 at 70K snp's:

37.88% ANE
11.19% ASE
33.92% WHG-UHG
1.91% East_Eurasian
0.00% West_African
0.00% East_African
15.11% ENF

So this in conjunction with the other sample does confirm elevated ANE and ASE in Afanasievo compared to Sintashta/Andronovo.

ArmandoR1b
06-11-2015, 09:55 PM
What is Genetiker's CT mean

Although the node that joins DE and CF is often referred to as
CT (in the sense that it encompasses the whole of haplogroups C to
T), and the clade that encompasses B, DE, and CF is often referred
to as BT (in the sense that it encompasses the whole of haplogroups
B to T), we are not in favor of these labels as they are inconsistent
with the meanings of other two-letter labels such as CF (referring to
the joint root of C and F, not the range C to F) and LT (referring to
the joint root of L and T, not the range L to T). Instead, we therefore
preferthemoreintuitivelabelsCDEFandBCDEF,respectiv ely,as
used in previous publications [Heinrich et al., 2009; Underhill and
Kivisild, 2007; van Oven et al., 2011].
Within the large K-M9 clade, which encompasses haplogroups
L, M, NO, P (including Q and R), S, and T, a deep bifurcation was
revealed, with haplogroups M, NO, P, and S falling in one subclade,
defined by marker M526 [Chiaroni et al., 2009], and haplogroups L
and T falling in the other subclade, defined by marker P326 [Mendez
et al., 2011]. The latter subclade logically being labeled LT [Mendez
et al., 2011], we “downgraded” the K label to designate the other,
M526-defined subclade, and used KLT for the unifying M9-defined
clade, as proposed previously [van Oven et al., 2011].

http://www.researchgate.net/publication/258117220_Seeing_the_Wood_for_the_Trees_A_Minimal_ Reference_Phylogeny_for_the_Human_Y_Chromosome

The person who creates the Y tree states CT as every haplogroup from C to T

also from 2013,.......... K is purely the M526 branch and K9 is the LT branch ..............that's how I read it.

It means that he is wrong. The specimens that he is showing to be CT, K and P1 are also positive for R1 but Genetiker isn't finding that. This isn't the first time Genetiker has posted bad information. He did it with a Hinxton sample also.

MasterRoshi
06-11-2015, 09:55 PM
The teal component in Yamnaya looks Caucasus derived so it makes sense that Sintashta was lower in it but this also means that Proto-Indo-Iranian ancestry among modern Indo-Iranians is not so high like many assume. I wonder which modern population is closest to Sintashta. It is also possible that Proto-Indo-Iranians mixed much with urban BMAC folks before they entered Afghanistan, India and Iran so that their EHG-like ancestry was already diluted in an early stage but they were quite successful in spreading their Y-DNA in Central and South Asia.

Sintashta is not Proto Indo_Iranian.

Proto Indo_Iranian might be Andronovo-BMAC but not even that is sure.

And I don't see any "teal component" here.

It is based on much lower Ks

The admixture results of this paper are not very accurate because they use Malta as refference because when they started with the research Haak results were not out yet so they didn't use any of the EHG as refference.

Andronovo appears in between Northeast Europeans and Pakistanis therefore geneticwise there is indeed a significant Proto Indo_Iranian ancestry in modern Indo_Iranian speakers.

MasterRoshi
06-11-2015, 10:29 PM
"Andronovo
Eurogenes ANE K7:
30.01% ANE
8.04% ASE
37.56% WHG-UHG
5.03% East_Eurasian
0.00% West_African
0.00% East_African
19.36% ENF"

Are you seeing any pattern? Andronovo and Sintashta look very similar.

Afanasievo as expected looks more Baikal oriented. Is there any reason to believe that Afanasievo has a Yamna origin? To me it looks it could also be the other way around.

"Afanasievo
ANE K7
48.33% ANE
10.31% ASE
27.49% WHG-UHG
0.00% East_Eurasian
0.00% West_African
0.00% East_African
13.87% ENF"

To me it looks like Eurogenes K7 is eating up some ENF ancestry as WHG-UHG.

Note WHG=/= WHG-UHG.

WHG-UHG is eating up the UHG ancestry in ENF. ENF in K7 is basically equivalent to basal eurasian.

We should use K8. Even Daviid uses that.

R.Rocca
06-11-2015, 10:39 PM
I will check into it in a few hours. If the script I ran had an issue with Z196 because it was an INDEL, or if it missed the U152 due to a low or heterozygous read.

Alex was right, this sample is U152+ Z196- Z56- PF6658- L2? Z36?

everest59
06-11-2015, 10:56 PM
Armenian 397 is giving me too much of a noise. Figures given the SNP count:

14.62% S-Indian
34.51% Baloch
41.86% Caucasian
5.53% NE-Euro
0.00% SE-Asian
0.91% Siberian
0.00% NE-Asian
0.00% Papuan
2.48% American
0.08% Beringian
0.00% Mediterranean
0.00% SW-Asian
0.00% San
0.00% E-African
0.00% Pygmy
0.00% W-African

I am doing 423 next, as it is a bigger file. Should get rid of the noise.

razyn
06-11-2015, 10:58 PM
Alex was right, this sample is U152+ Z196- Z56- PF6658- L2? Z36?

So, only DF27's brother clade below ZZ11_1 (if Alex is also right about that). I'll take it; but whether his P312 brother DF27 was yet unborn -- or just elsewhere -- is not addressed by that sample. Btw, is there available dating info (from the archaeological context of the sample, not the guesstimate du jour about the age of the youngest detected SNP)? "Bell Beaker" has a fairly broad date range.

nuadha
06-11-2015, 11:56 PM
Anything can happen, I guess, but a lot depends on whether you are talking about y-dna or mtDNA or both, what clades you have in mind, and the ages of the samples.

I could be wrong, but I don't think we are going to see a pattern of general European population continuity, because that would not make sense in light of the tremendous changes that occurred with the onset of the Late Neolithic/Bronze Age period. Among other things, somehow a big shot of ANE was introduced into the population of Europe and it began speaking Indo-European languages.

But I suspect even after every last spadeful of Iberian and French soil is turned, there will still be some who believe that Mesolithic Western European R1b is still out there somewhere waiting to be unearthed.

The question remains as to why the yamnaya took western europrean graves back to the steppe and buried them with yamnaya ornaments. I guess it could just as well be the fault of western european r1b folk who cremated their dead western europe but would often start families in russia where they would only cremate the ones that migrated, then their long term descendants much have reached a dead end unless then moved their family back to western europe.

rock hunter
06-12-2015, 01:11 AM
Ancient DNA reveals how Europeans developed light skin and lactose tolerance
Daniel Zadik, The Conversation
June 11, 2015

Food intolerance is often dismissed as a modern invention and a "first-world problem". However, a study analysing the genomes of 101 Bronze-Age Eurasians reveals that around 90% were lactose intolerant.

The research also sheds light on how modern Europeans came to look the way they do – and that these various traits may originate in different ancient populations. Blue eyes, it suggests, could come from hunter gatherers in Mesolithic Europe (10,000 to 5,000 BC), while other characteristics arrived later with newcomers from the East.

About 40,000 years ago, after modern humans spread from Africa, one group moved north and came to populate Europe as well as north, west and central Asia. Today their descendants are still there and are recognisable by some very distinctive characteristics. They have light skin, a range of eye and hair colours and nearly all can happily drink milk.

However, exactly when and where these characteristics came together has been anyone's guess. Until now.

Clash of cultures

Throughout history, there has been a pattern of cultures rising, evolving and being superseded. Greek, Roman and Byzantine cultures each famously had their 15 minutes as top dog. And archaeologists have defined a succession of less familiar cultures that rose and fell before that, during the Bronze Age. So far it has been difficult to work out which of these cultures gave rise to which – and eventually to today's populations.

The Bronze Age (around 3,000–1,000 BC) was a time of major advances, and whenever one culture developed a particularly advantageous set of technologies, they become able to support a larger population and to dominate their neighbours. The study found that the geographical distributions of genetic variations at the beginning of the Bronze Age looked very different to today's, but by the end it looked pretty similar, suggesting a level of migration and replacement of peoples not seen in western Eurasia since.
Ancient DNA reveals how Europeans developed light skin and lactose tolerance
Yamnaya skull. Credit: Natalia Shishlina

One people that was particularly important in the spread of both early Bronze-Age technologies and genetics were the Yamnaya. With a package of technologies including the horse and the wheel, they exploded out of the Russian and Ukrainian Steppe into Europe, where they met the local Neolithic farmers.

By comparing DNA from various Bronze-Age European cultures to that of both Yamnaya and the Neolithic farmers, researchers found that most had a mixture of the two backgrounds. However the proportions varied, with the Corded Ware people of northern Europe having the highest proportion of Yamnaya ancestry.

And it appears that the Yamnaya also moved east. The Afanasievo culture of the Altai-Sayan region in central Asia seemed to be genetically indistinguishable from the Yamnaya, suggesting a colonisation with little or no interbreeding with pre-existing populations.

Mutations traced

So how have traits that were rare or non-existent in our African ancestors come to be so common in western Eurasia?

The DNA of several hunter gatherers living in Europe long before the Bronze Age was also tested. It showed that they probably had a combination of features quite striking to the modern eye: dark skin with blue eyes.
Ancient DNA reveals how Europeans developed light skin and lactose tolerance
Reconstruction of a Yamnaya person from the Caspian steppe in Russia about 5,000-4,800 BC.

The blue eyes of these people – and of the many modern Europeans who have them – are thanks to a specific mutation near a gene called OCA2. As none of the Yamnaya samples have this mutation, it seems likely that modern Europeans owe this trait to their ancestry from these European hunter gatherers of the Mesolithic (10,000-5,000 BC).

Two mutations responsible for light skin, however, tell quite a different story. Both seem to have been rare in the Mesolithic, but present in a large majority by the Bronze Age (3,000 years later), both in Europe and the steppe. As both areas received a significant influx of Middle Eastern farmers during this time, one might speculate that the mutations arose in the Middle East. They were probably then driven to high levels by natural selection, as they allowed the production of sufficient vitamin D further north despite relatively little sunlight, and/or better suited people to the new diet associated with farming.

Another trait that is nearly universal in modern Europeans (but not around the world) is the ability to digest the lactose in milk into adulthood. As cattle and other livestock have been farmed in western Eurasia since long before, one might expect such a mutation to already be widespread by the Bronze Age. However the study revealed that the mutation was found in around 10% of their Bronze Age samples.

Interestingly, the cultures with the most individuals with this mutation were the Yamnaya and their descendents. These results suggest that the mutation may have originated on the steppe and entered Europe with the Yamnaya. A combination of natural selection working on this advantageous trait and the advantageous Yamnaya culture passed down alongside it could then have helped it spread, although this process still had far to go during the bronze age.

This significant study has left us with a much more detailed picture of Bronze Age Europeans: they had the light skin and range of eye colours we know today. And although most would have got terrible belly ache from drinking milk, the seeds for future lactose tolerance were sown and growing.


http://phys.org/news/2015-06-ancient-dna-reveals-europeans-skin.html#jCp

rock hunter
06-12-2015, 01:15 AM
DNA Deciphers Roots of Modern Europeans
JUNE 10, 2015

Photo of
A male skeleton associated with the Yamnaya culture near Samara, Russia. Credit Pavel Kuznetsov



For centuries, archaeologists have reconstructed the early history of Europe by digging up ancient settlements and examining the items that their inhabitants left behind. More recently, researchers have been scrutinizing something even more revealing than pots, chariots and swords: DNA.

On Wednesday in the journal Nature, two teams of scientists — one based at the University of Copenhagen and one based at Harvard University — presented the largest studies to date of ancient European DNA, extracted from 170 skeletons found in countries from Spain to Russia. Both studies indicate that today’s Europeans descend from three groups who moved into Europe at different stages of history.

The first were hunter-gatherers who arrived some 45,000 years ago in Europe. Then came farmers who arrived from the Near East about 8,000 years ago.

Finally, a group of nomadic sheepherders from western Russia called the Yamnaya arrived about 4,500 years ago. The authors of the new studies also suggest that the Yamnaya language may have given rise to many of the languages spoken in Europe today.

Photo
A Yamnaya skull found near Samara, Russia, colored with ocher. Yamnaya expansion into Europe was most likely relatively peaceful. Credit Natalia Shishlina
Ron Pinhasi, an archaeologist at University College Dublin who was not involved in either study, said that the new studies were “a major game-changer. To me, it marks a new phase in ancient DNA research.”

The two teams worked independently, studying different skeletons and using different methods to analyze their DNA.

The Harvard team collected DNA from 69 human remains dating back 8,000 years and cataloged the genetic variations at almost 400,000 different points. The Copenhagen team collected DNA from 101 skeletons dating back about 3,400 years and sequenced the entire genomes.

Both teams also compared the newly sequenced DNA to genes retrieved from other ancient Europeans and Asians, and to living humans.

Until about 9,000 years ago, Europe was home to a genetically distinct population of hunter-gatherers, the researchers found. Then, between 9,000 and 7,000 years ago, the genetic profiles of the inhabitants in some parts of Europe abruptly changed, acquiring DNA from Near Eastern populations.

Archaeologists have long known that farming practices spread into Europe at the time from Turkey. But the new evidence shows that it wasn’t just the ideas that spread — the farmers did, too.

The hunter-gatherers didn’t disappear, however. They managed to survive in pockets across Europe between the farming communities.

“It’s an amazing cultural process,” said David Reich, a geneticist at Harvard Medical School who led the university’s team. “You have groups which are as genetically distinct as Europeans and East Asians. And they’re living side by side for thousands of years.”

Between 7,000 and 5,000 years ago, however, hunter-gatherer DNA began turning up in the genes of European farmers. “There’s a breakdown of these cultural barriers, and they mix,” said Dr. Reich.

About 4,500 years ago, the final piece of Europe’s genetic puzzle fell into place. A new infusion of DNA arrived — one that is still very common in living Europeans, especially in central and northern Europe.


The closest match to this new DNA, both teams of scientists found, comes from skeletons found in Yamnaya graves in western Russia and Ukraine.

Archaeologists have long been fascinated by the Yamnaya, who left behind artifacts on the steppes of western Russia and Ukraine dating from 5,300 to 4,600 years ago. The Yamnaya used horses to manage huge herds of sheep, and followed their livestock across the steppes with wagons full of food and water.

It was an immensely successful way of life, allowing the Yamnaya to build huge funeral mounds for their dead, which they filled with jewelry, weapons and even entire chariots.

Photo
A skeleton buried by a Middle Neolithic culture found near Saxony-Anhalt, Germany. A review of DNA from skeletons across Europe indicated that today's Europeans are descended from three groups who moved there at different stages of history. Credit Juraj Lipták/LDA Sachsen-Anhalt
David W. Anthony, an archaeologist at Hartwick College and a co-author on the Harvard study, said it was likely that the expansion of Yamnaya into Europe was relatively peaceful. “It wasn’t Attila the Hun coming in and killing everybody,” he said.

Instead, Dr. Anthony thought the most likely scenario was that the Yamnaya “entered into some kind of stable opposition” with the resident Europeans that lasted for a few centuries. But then gradually the barriers between the cultures eroded.

The Copenhagen team’s study suggests that the Yamnaya didn’t just expand west into Europe, however. The scientists examined DNA from 4,700-year-old skeletons from a Siberian culture called the Afanasievo. It turns out that they inherited Yamnaya DNA, too.

Dr. Anthony was surprised by the possibility that Yamnaya pushed out over a range of about 4,000 miles. “I myself have a hard time wrapping my head around explanations for that,” he said.

The two studies also add new fuel to a debate about how languages spread across Europe and Asia. Most European tongues belong to the Indo-European family, which also incudes languages in southern and Central Asia.

For decades, linguists have debated how Indo-European got to Europe. Some favor the idea that the original farmers brought Indo-European into Europe from Turkey. Others think the language came from the Russian steppes thousands of years later.

The new genetic results won’t settle the debate, said Eske Willerslev, an evolutionary biologist at Copenhagen University who led the Danish team. But he did think the results were consistent with the idea that the Yamnaya brought Indo-European from the steppes to Europe.

The eastward expansion of Yamnaya, evident in the genetic findings, also supports the theory, Dr. Willerslev said. Linguists have long puzzled over an Indo-European language once spoken in western China called Tocharian. It is only known from 1,200-year-old manuscripts discovered in ancient desert towns. It is possible that Tocharian was a vestige of the eastern spread of the Yamnaya.

“We can just say that the expansion fits very well with the geographical spread of the Indo-European language,” said Dr. Willerslev.

Paul Heggarty, a linguist at the Max Planck Institute of Evolutionary Anthropology, said that the new studies were important, but were still too limited to settle the debate over the origins of Indo-European. “I don’t think we’re there yet,” he said.

Dr. Heggarty noted that the studies showed the arrival of Yamnaya in Central Europe about 4,500 years ago. But Greek is an Indo-European language, and the oldest evidence of writing in Europe shows that Greek had developed about 3,500 years ago. By then, it was distinct from other Indo-European languages in Southern Europe, like Latin.

If the Yamnaya were the source of Indo-European languages, they would have had to get to southern Europe soon after they made it to Central Europe.

Dr. Heggarty speculated instead that early European farmers, the second wave of immigrants, may have brought Indo-European to Europe from the Near East. Then, thousands of years later, the Yamnaya brought the language again to Central Europe.

More ancient DNA could swing the balance of evidence in favor of one theory over the other, Dr. Heggarty said. A stronger case for a steppe origin of Indo-European might emerge, for example, if scientists discovered that Greeks around 4,500 years ago abruptly acquired Yamnaya DNA.

“Let’s see whether they look like the steppe people or not,” he said.
http://www.nytimes.com/2015/06/16/science/dna-deciphers-roots-of-modern-europeans.html?_r=0

Dr_McNinja
06-12-2015, 02:08 AM
To me it looks like Eurogenes K7 is eating up some ENF ancestry as WHG-UHG.

Note WHG=/= WHG-UHG.

WHG-UHG is eating up the UHG ancestry in ENF. ENF in K7 is basically equivalent to basal eurasian.

We should use K8. Even Daviid uses that.

Yeah K7 has a lopsided WHG:ENF ratio compared to K8. We can't use K8, that was the one used for crowdfunding and wasn't released publicly like K7.

ADW_1981
06-12-2015, 02:51 AM
Are any of these samples on Gedmatch yet? If so, what are the kit #s? I'm a little curious to see how they fare on the Eurogenes K=15 calculator. I did not read the Allentoft paper top to bottom, but is it true the Eurasian reference is Ma'lta rather than the R1a/R1b Russian hunter gatherers?

MasterRoshi
06-12-2015, 02:58 AM
Armenian 397 is giving me too much of a noise. Figures given the SNP count:

14.62% S-Indian
34.51% Baloch
41.86% Caucasian
5.53% NE-Euro
0.00% SE-Asian
0.91% Siberian
0.00% NE-Asian
0.00% Papuan
2.48% American
0.08% Beringian
0.00% Mediterranean
0.00% SW-Asian
0.00% San
0.00% E-African
0.00% Pygmy
0.00% W-African

I am doing 423 next, as it is a bigger file. Should get rid of the noise.
Just like RISE416 no Red Sea/Southwest Asian component. Seems really like This component was brought to Mesopotamia and Transcaucasus vie the Assyrians. And this must also be the component the people were talking about when saying " a Middle Eastern and North African specific gene slowly replacing the Early farmer DNA."

It is logical that Semites would have reached Anatolia before the Transcaucasus and Mesopotamia, since it is slightly closer and easier to reach.

The appearance of Red Sea component seems to be connected to the expansion of Afro:Asiatic language family.

Coldmountains
06-12-2015, 03:09 AM
Sintashta is not Proto Indo_Iranian.

Proto Indo_Iranian might be Andronovo-BMAC but not even that is sure.

And I don't see any "teal component" here.

It is based on much lower Ks

The admixture results of this paper are not very accurate because they use Malta as refference because when they started with the research Haak results were not out yet so they didn't use any of the EHG as refference.

Andronovo appears in between Northeast Europeans and Pakistanis therefore geneticwise there is indeed a significant Proto Indo_Iranian ancestry in modern Indo_Iranian speakers.

You are probably right about Sintashta, I got a wrong impression when I saw the results at first they are indeed south central Asian shifted in some way and just one or two admixture events away from the Proto-Indo-Iranians who directly contributed to our ancestry. The paper exaggerates the connection to Corded Ware in my opinion but Sintashta/Andronovo have more MN European ancestry than Yamnaya it seems. The Andronovo samples were from the most northern and eastern zone of the Andronovo horizon so they show some east Eurasian admixture but also higher south central asian-like admixture than Sintashta. I guess we need Andronovo samples from the south to estimate direct Andronovo ancestry among modern Indo-Iranians. Testing BMAC would be also very important.

MasterRoshi
06-12-2015, 03:16 AM
You are probably right about Sintashta, I got a wrong impression when I saw the results at first they are indeed south central Asian shifted in some way and just one or two admixture events away from the Proto-Indo-Iranians who directly contributed to our ancestry. The paper exaggerates the connection to Corded Ware in my opinion but Sintashta/Andronovo have more MN European ancestry than Yamnaya it seems. The Andronovo samples were from the most northern and eastern zone of the Andronovo horizon so they show some east Eurasian admixture but also higher south central asian-like admixture than Yamnaya. I guess we need Andronovo samples from the south to estimate direct Andronovo ancestry among modern Indo-Iranians.


autosomal results of Iron Age Iranic tribes such as Scythians, Cimmerians, Medes, Persian and Massagaeta would also be important to see if Andronovo_BMAC is really a good fit for Indo_Iranian origin. It wouldn't be much suprising for me when it turned out that Andronovo is just another Indo European culture which died out. And ancestors of the Indo_Iranians came from the same region as the ancestors of proto_Greeks, Thracians and Armenians.

Christina
06-12-2015, 03:24 AM
Ancient DNA reveals how Europeans developed light skin and lactose tolerance
Daniel Zadik, The Conversation
June 11, 2015


The research also sheds light on how modern Europeans came to look the way they do – and that these various traits may originate in different ancient populations. Blue eyes, it suggests, could come from hunter gatherers in Mesolithic Europe (10,000 to 5,000 BC), while other characteristics arrived later with newcomers from the East.

About 40,000 years ago, after modern humans spread from Africa, one group moved north and came to populate Europe as well as north, west and central Asia. Today their descendants are still there and are recognisable by some very distinctive characteristics. They have light skin, a range of eye and hair colours and nearly all can happily drink milk.

However, exactly when and where these characteristics came together has been anyone's guess. Until now.

Clash of cultures

Throughout history, there has been a pattern of cultures rising, evolving and being superseded. Greek, Roman and Byzantine cultures each famously had their 15 minutes as top dog. And archaeologists have defined a succession of less familiar cultures that rose and fell before that, during the Bronze Age. So far it has been difficult to work out which of these cultures gave rise to which – and eventually to today's populations.




Mutations traced

So how have traits that were rare or non-existent in our African ancestors come to be so common in western Eurasia?

The DNA of several hunter gatherers living in Europe long before the Bronze Age was also tested. It showed that they probably had a combination of features quite striking to the modern eye: dark skin with blue eyes.
Ancient DNA reveals how Europeans developed light skin and lactose tolerance
Reconstruction of a Yamnaya person from the Caspian steppe in Russia about 5,000-4,800 BC.

The blue eyes of these people – and of the many modern Europeans who have them – are thanks to a specific mutation near a gene called OCA2. As none of the Yamnaya samples have this mutation, it seems likely that modern Europeans owe this trait to their ancestry from these European hunter gatherers of the Mesolithic (10,000-5,000 BC).

Two mutations responsible for light skin, however, tell quite a different story. Both seem to have been rare in the Mesolithic, but present in a large majority by the Bronze Age (3,000 years later), both in Europe and the steppe. As both areas received a significant influx of Middle Eastern farmers during this time, one might speculate that the mutations arose in the Middle East. They were probably then driven to high levels by natural selection, as they allowed the production of sufficient vitamin D further north despite relatively little sunlight, and/or better suited people to the new diet associated with farming.

Another trait that is nearly universal in modern Europeans (but not around the world) is the ability to digest the lactose in milk into adulthood. As cattle and other livestock have been farmed in western Eurasia since long before, one might expect such a mutation to already be widespread by the Bronze Age. However the study revealed that the mutation was found in around 10% of their Bronze Age samples.

Interestingly, the cultures with the most individuals with this mutation were the Yamnaya and their descendents. These results suggest that the mutation may have originated on the steppe and entered Europe with the Yamnaya. A combination of natural selection working on this advantageous trait and the advantageous Yamnaya culture passed down alongside it could then have helped it spread, although this process still had far to go during the bronze age.




http://phys.org/news/2015-06-ancient-dna-reveals-europeans-skin.html#jCp

Thanks for posting the text from this article. It has some good points and some questionable.

1. I would hardly call the "Greeks, Romans, and Byzantines" as having "15 minutes of fame." Talk about a modern phrase (like the word "literally") coming to mean something entirely different...

2. If I am understanding the synthesis then, a modern European who is light-skinned, blonde-haired, and blue eyed, acquired these traits from various groups. If this summary in the article is accurate, then basically the blue-eyed gene was likely a WHG thing, light-skin was a Yamnaya thing perhaps (and there is no uniformity on when blonde hair arose).

3. The almost total lack of lactose tolerance in these samples is consistent with a model of what I and others have been posting about for years. It is likely there was a HUGE expansion from East to West, carrying the IE languages and new genes. BUT, there was a SECOND expansion say 500 years later, out of Iberia or even England, bringing the huge numbers of lactose tolerance we now see, which likely spread west to east.

royking
06-12-2015, 03:25 AM
The Bronze Age Hungarian (RISE374) who is Haplogroup G is derived for Z6308 which is equivalent to PF3177--the immediate SNP above Oetzi's L91 Y, according to Ted Kandell of Open Genomes Foundation.

R.Rocca
06-12-2015, 03:50 AM
German Corded Ware sample RISE434 (2880-2630 BC) is at least R1b1a2. He has two known M269 equivalents in PF6508 and PF6527 and is also positive for Y429, which Thomas Krahn has as the same level as L23. I'll try to keep looking at the sample to see if anything else comes up downstream.

parasar
06-12-2015, 04:05 AM
autosomal results of Iron Age Iranic tribes such as Scythians, Cimmerians, Medes, Persian and Massagaeta would also be important to see if Andronovo_BMAC is really a good fit for Indo_Iranian origin. It wouldn't be much suprising for me when it turned out that Andronovo is just another Indo European culture which died out. And ancestors of the Indo_Iranians came from the same region as the ancestors of proto_Greeks, Thracians and Armenians.

If that is the case, and I have seen some support for that proposition, then it is not Z93 that is Indo-Iranian-proto_Greeks-Thracians-Armenian marker but Y-J and Y-G.

Then we would have to bring Z93 into South Asia very late. In that case even Y-R1b is not a Proto-Indo-European marker but became Indo-Europeanized in the Indo-Iranian-proto_Greeks-Thracians-Armenian-steppe contact zone - Yamna. This then transferred to Z645 in the steppe-forest zone with the Z283 prong going to Corded Ware and the Z93 prong (the Shak-Scythian types) merging into the Indo-Iranians.

DMXX
06-12-2015, 04:21 AM
Sample RISE495 (Karasuk culture (http://en.wikipedia.org/wiki/Karasuk_culture), local post-Andronovo offshoot in East-Central Asia) came out mtDNA D4j1. Their mutations:



73G 263G 489C 750G 1438G 2706G 3010A 3107T 4263G 4883T 5262A 7028T 8414T 9540C 9948A 10398G 10400T 10873C 11696A 11719A 12705T 14130T 14668T 14766T 14769G 14783C 15043A 15301A 15326G 16223T 16260T 16362C


My HVR1+2 results (differences in red, additions +):



16223T 16362C 16519C 73G 263G 309.1C 315.1C 489C


Excluding 309.1 and 315.1, full match as far as the other polymorphism sites are concerned.* I might just be looking at the results of my distant maternal ancestor...

* Predicted D4j5 via James Lick's mtDNA predictor, therefore not confirmed yet

Hando
06-12-2015, 04:45 AM
Does anyone have access to the paper?
Hi, yes I do. And if anyone needs access to any papers I can get it through my UK academic institution. So just let me know. I can also email it as a PDF.
I'm getting through this thread slowly, so just PM me.

R.Rocca
06-12-2015, 04:49 AM
German Corded Ware sample RISE434 (2880-2630 BC) is at least R1b1a2. He has two known M269 equivalents in PF6508 and PF6527 and is also positive for Y429, which Thomas Krahn has as the same level as L23. I'll try to keep looking at the sample to see if anything else comes up downstream.

And an even bigger development...the sample is also Z2103- and also PF6543- (an L11 equivalent). Unfortunately no read at L51 or equivalent SNPs. So, if this is an equivalent M269+L23+Z2103-L11- sample, it could very well be "the" sample R1b folks have been waiting for for years. The upper range age of this sample makes it several hundred years older than any Bell Beaker in Bavaria and about contemporary with the absolute oldest Bell Beaker materials found anywhere. The site's location in Bavaria one mile from the Danube makes it an even more intriguing sample.

Krefter
06-12-2015, 05:58 AM
And an even bigger development...the sample is also Z2103- and also PF6543- (an L11 equivalent). Unfortunately no read at L51 or equivalent SNPs. So, if this is an equivalent M269+L23+Z2103-L11- sample, it could very well be "the" sample R1b folks have been waiting for for years. The upper range age of this sample makes it several hundred years older than any Bell Beaker in Bavaria and about contemporary with the absolute oldest Bell Beaker materials found anywhere. The site's location in Bavaria one mile from the Danube makes it an even more intriguing sample.

Can you place your work with Y DNA Allentoft in some-type of document/file/spreadsheet? It's hard to keep track of your work.

You could do something similar to this.

https://docs.google.com/spreadsheets/d/1HuNPykGuq2PbHkUOL5dCiwrveIy-OGO2qOklwfsayW8/edit#gid=2120948378

Krefter
06-12-2015, 06:08 AM
And an even bigger development...the sample is also Z2103- and also PF6543- (an L11 equivalent). Unfortunately no read at L51 or equivalent SNPs. So, if this is an equivalent M269+L23+Z2103-L11- sample, it could very well be "the" sample R1b folks have been waiting for for years. The upper range age of this sample makes it several hundred years older than any Bell Beaker in Bavaria and about contemporary with the absolute oldest Bell Beaker materials found anywhere. The site's location in Bavaria one mile from the Danube makes it an even more intriguing sample.

What about Bell Beaker R1b? Isn't it so far 2/2 R1b-U152?

David Mc
06-12-2015, 06:17 AM
What about Bell Beaker R1b? Isn't it so far 2/2 R1b-U152?

I think the significance of this is that it is that it is an L11 equivalent, it is earlier (in Bavaria when the very earliest Bell Beakers are coming into being), and it is close to the Danube, which is at least suggestive of an east-west trajectory.

MasterRoshi
06-12-2015, 06:20 AM
Armenian 397 is giving me too much of a noise. Figures given the SNP count:

14.62% S-Indian
34.51% Baloch
41.86% Caucasian
5.53% NE-Euro
0.00% SE-Asian
0.91% Siberian
0.00% NE-Asian
0.00% Papuan
2.48% American
0.08% Beringian
0.00% Mediterranean
0.00% SW-Asian
0.00% San
0.00% E-African
0.00% Pygmy
0.00% W-African

I am doing 423 next, as it is a bigger file. Should get rid of the noise.


This particular result (if correct) here looks more Iranian than anything else and is even slightly more eastern and less Southern(Southwest Asian) shifted than modern Iranians!

RISE416 sample from Bronze Age Armenia appears roughly like an even mix of Basque and Adygei/Lezgians.

David Mc
06-12-2015, 06:34 AM
I think the significance of this is that it is that it is an L11 equivalent, it is earlier (in Bavaria when the very earliest Bell Beakers are coming into being), and it is close to the Danube, which is at least suggestive of an east-west trajectory.

Put another way, R1b wasn't sitting around in Portugal when the Bell Beaker cultural package came into being; it was in modern day Bavaria, not too far away from where we find other R1b Beaker people a hundred years later. Their is a continuity of some sort going on here.

DMXX
06-12-2015, 06:34 AM
This particular result (if correct) here looks more Iranian than anything else and is even slightly more eastern and less Southern(Southwest Asian) shifted than modern Iranians!

RISE416 sample from Bronze Age Armenia appears roughly like an even mix of Basque and Adygei/Lezgians.

If the S. Indian and SW Asian scores were swapped round, this would be remarkably similar with the Behar et al. Iranian dataset results at Harappa.

We'll have to see how the official genotype files look, whenever they're published. I suspect a lot of the inferences being made in this thread won't hold up.

The uniparental marker results (http://www.nature.com/nature/journal/v522/n7555/fig_tab/nature14507_SF6.html) are quite interesting. Yamnaya appears to be largely defined by mtDNA T2 and U5. On the other hand, there's a surprisingly high frequency of mtDNA J2 among Afanasievo. I'm also surprised to see Afanasievo was, as far as this study permits, fully West Eurasian maternally (J2, T2, U5). If the archaeological data is anything to go by, I'd expect the likes of mtDNA A, C, D and/or Z to make an appearance with more samples in keeping with previous data across the steppe zone.

Radboud
06-12-2015, 06:50 AM
R1b U106 is yet to be found....