PDA

View Full Version : aDNA from Early Bronze Age Cultures of the North Pontic Steppe



Jean M
09-18-2014, 06:55 PM
Jeff Pashnick Thesis: Genetic Analysis of Ancient Human Remains from the Early Bronze Age Cultures of the North Pontic Steppe Region
http://scholarworks.gvsu.edu/theses/737/
Full content under embargo until September 08, 2017.


We extracted ancient mtDNA from remains of 11 individuals belonging to late Neolithic and EBA populations of the NPR. Using single nucleotide polymorphisms (SNPs) as markers, we established mtDNA haplogroups of nine out of 11 individuals.

There follows a huge weight of analysis for such a small sample, but here is it for what it is worth:


We found genetic evidence through mtDNA haplogroup frequencies and PCA linking the Catacomb people to hunter-gatherer populations from northern Europe and Russia. On the other hand, data on mtDNA haplogroup frequencies of individuals from the Yamna culture associated them with farming and pastoralist type populations from southwest and central Europe. An FST analysis of mtDNA haplogroup frequency distribution showed that the Yamna are most closely related to the Boyko group of ethnic Carpathian highlanders than to other modern European groups used in the study. The Catacomb people appeared genetically different from all other population groups in the FST analysis, including the Yamna group, challenging the current understanding of the relationship between the Yamna and Catacomb populations. Further statistical analysis using an exact test of population differentiation confirmed genetic differences in mtDNA haplogroup frequencies between Yamna and Catacomb. The exact test also revealed a lack of genetic differentiation between the Yamna and the modern Ukrainian population, as well as Lemko, another group of Carpathian highlanders.

GailT
09-18-2014, 08:57 PM
Apparently there was a presentation available two years ago (link (http://dienekes.blogspot.com/2012/10/mtdna-of-bronze-age-pastoral-nomads-of.html)) but no longer available. I suppose one benefit of keeping the results secret for 5 years is that no one can question your conclusions.

Jean M
09-18-2014, 09:05 PM
I imagine that the embargo is simply to protect planned publication of these results.

The conclusions don't strike me as particularly bizarre. A genetic difference between Yamna and Catacomb people would fit my current idea that the dominant input into the latter was Iranian-speaking nomads from the Asian steppe, who might look somewhat different from Yamna if they had acquired some wives from Central Asia. But a lot depends on which site within the wide Yamna Horizon they actually sampled. I suspect one towards the west (heavily Cucuteni in terms of mtDNA), which might well exaggerate the difference between Yamna and Catacomb. But frankly a few mtDNA samples won't get us real answers. We need ancient Y-DNA and preferably full genomes as well.

nuadha
09-19-2014, 04:07 AM
Apparently there was a presentation available two years ago (link (http://dienekes.blogspot.com/2012/10/mtdna-of-bronze-age-pastoral-nomads-of.html)) but no longer available. I suppose one benefit of keeping the results secret for 5 years is that no one can question your conclusions.

Do they realize how fast the scene is changing? Im guessing sandra wilde did the same sort of stalling which made her results look very unimpressive at the time she published.

Jean M
09-19-2014, 07:13 AM
Do they realize how fast the scene is changing? Im guessing sandra wilde did the same sort of stalling which made her results look very unimpressive at the time she published.

It is not a question of stalling. The problem lies in the academic process. A PhD takes three years or more. A PhD project with aDNA designed in 2010 or thereabouts would focus on mtDNA. A lot of research is done as PhD projects. So we are still getting the results now of projects set up years ago. Inevitably it is disappointing by comparison with the Y-DNA and full genome studies being done at the cutting edge of the science.

GailT
09-19-2014, 01:18 PM
The problem lies in the academic process. A PhD takes three years or more. A PhD project with aDNA designed in 2010 or thereabouts would focus on mtDNA.

It is normal to protect results until you have an opportunity to publish, but a 3 year embargo after defense of the dissertation is extreme. Additionally, in this case, there are no details provided on what haplogroups were found. It is possible to provide some information that would allow evaluation of the results without jeopordizing future publications. So it's the combination of long embargo and minimal information that it is disturbing here, especially given that this is a quickly evolving field in which there has been some questionable scholarship.

Jean M
09-19-2014, 08:43 PM
This paper from the same lab may be helpful, since Pashnick says the Yamna were like the Boyko and Lemko.

Nikitin AG et al., Mitochondrial DNA sequence variation in the Boyko, Hutsul, and Lemko populations of the Carpathian highlands, Hum Biol. 2009 Feb . 81(1):43-58. http://www.lemko.org/scholar/DNA.pdf

T101
09-19-2014, 11:17 PM
This paper from the same lab may be helpful, since Pashnick says the Yamna were like the Boyko and Lemko.

That's interesting and it does makes some sense. The Boykos, Hutsuls, and Lemkos, were probably descended from the White Croats, or Rus and therefore have significant ancestry from the Yamnaya horizon. They were driven from those areas of Western Ukraine by the Mongols or Turks (Pechenegs,) and sought safety in the Western Carpathians, where they remained relatively isolated and had little intermixing with other groups, so that today, they still maintain a high proportion of the early farming and pastoralist type mtDna (H, J, K, T2 etc.)

While the Catacomb culture was an intrusive Indo-Iranian group from further east in Kazakhstan and probably has a much higher percentage of Mesolithic European hunter-gatherer mtdna (U4, U5, H1, H3, C4 - thus the Northern European connection,) and in addition some Asian mtdna.

Hando
09-21-2014, 05:09 AM
The Boykos, Hutsuls, and Lemkos, were probably descended from the White Croats, or Rus and therefore have significant ancestry from the Yamnaya horizon ... they still maintain a high proportion of the early farming and pastoralist type mtDna (H, J, K, T2 etc.)

While the Catacomb culture was an intrusive Indo-Iranian group from further east in Kazakhstan and probably has a much higher percentage of Mesolithic European hunter-gatherer mtdna (U4, U5, H1, H3, C4 - thus the Northern European connection,) and in addition some Asian mtdna.
1)I thought Yamnaya was IE and therefore would not have such high frequencies of early farming mtDNA.
2)If Catacomb was an intrusive Indo Iranian group from Kazahkstan how is it possible that they have a much higher percentage of Mesolithic European hunter gatherer mtDNA? Because IE migrated to Kazahkstan earlier from Europe?

Jean M
09-21-2014, 09:17 AM
I thought Yamnaya was IE and therefore would not have such high frequencies of early farming mtDNA.

How can this paper establish the actual frequencies of mtDNA haplogroups within the entire Yamnaya Horizon (which stretched all the way across the European steppe) if they only sampled a tiny number of individuals, probably from just one site? That is what I meant by complaining about the huge weight of inference from a tiny, weeny sample.

Until we have the full details of this thesis, we don't know exactly how many Yamna individuals were sampled, but the total results from Yamna and Catacomb was nine. So maybe they got four or five Yamna results. It seems likely that the actual frequencies of farmer mtDNA varied across the steppe, with higher levels in the centre and west than the east.


If Catacomb was an intrusive Indo Iranian group from Kazahkstan how is it possible that they have a much higher percentage of Mesolithic European hunter gatherer mtDNA? Because IE migrated to Kazahkstan earlier from Europe?

Yes indeed. Iranian is part of the IE family.

Jean M
09-21-2014, 09:19 AM
Mesolithic European hunter-gatherer mtdna (U4, U5, H1, H3, C4)

U4 and U5 are certainly European hunter-gatherer signatures. H1 and H3 appear in early farmer samples. C4 is central Asian.

T101
09-21-2014, 04:49 PM
U4 and U5 are certainly European hunter-gatherer signatures. H1 and H3 appear in early farmer samples. C4 is central Asian.

I completely agree. Thank you for correcting my hasty post. However, there may be a few more corrections still required! Maybe I can get to revising it next w/e.

GailT
09-21-2014, 06:02 PM
Thanks for the link and here is the summary of the Boyko and Lemko samples, which they describe as being similar to the ancient Yamnaya samples.


The Boyko sample analysis showed atypically low haplogroup H frequency (20%) for a European population. Boyko samples showed no H subhaplogroup subdivision to the extent of the resolution capacity of the current study (Table 2). Boykos displayed the highest haplogroup T frequency (30%)

The Lemko sample analysis showed the second lowest haplogroup H frequency (32.1%) in the region after Boykos. Among the 17 haplogroup H samples in Lemko, 4 (23.5% of the haplogroup H samples, 7.5% overall) were Hl and 1 (5.9%, 1.9% overall) was H2 (Table 2). The Lemko sample also contained the highest frequency of haplogroup I (11.3%) in Europe, identical to that of the population of Krk Island (Croatia) in the Adriatic Sea (Pericic et al. 2005). Three samples (5.7%) in the Lemko group belonged to the M* clade, similar to the number of M* specimens uncovered in a subset of the Hungarian population (5.5%) in a recent study

So we can probably expect mtDNA H and T in the ancient Yamnaya samples, and perhaps I, J and M*. (For comparison, from Jean's Ancient DNA page, the Dnieper-Donets of about 5200 BCE had mtDNA H, C, T, U3, U5.)

So the Yamnaya (and Proto-Indo Europeans) might be part of the Neolithic revolution, heavily influenced by pastoralists from the Near East, perhaps with a smaller influence from northern foragers from eastern Europe and central Asia. This would make it difficult to use mtDNA to distinguish Neolithic European farmers from later Indo European migrations if they have similar Neolithic mtDNA ancestry.

Táltos
09-21-2014, 06:24 PM
Thanks for the link and here is the summary of the Boyko and Lemko samples, which they describe as being similar to the ancient Yamnaya samples.



So we can probably expect mtDNA H and T in the ancient Yamnaya samples, and perhaps I, J and M*. (For comparison, from Jean's Ancient DNA page, the Dnieper-Donets of about 5200 BCE had mtDNA H, C, T, U3, U5.)

So the Yamnaya (and Proto-Indo Europeans) might be part of the Neolithic revolution, heavily influenced by pastoralists from the Near East, perhaps with a smaller influence from northern foragers from eastern Europe and central Asia. This would make it difficult to use mtDNA to distinguish Neolithic European farmers from later Indo European migrations if they have similar Neolithic mtDNA ancestry.
It is too bad they could not distinguish the M* in the Lemkos. I find it interesting as I have a matches that are either M1a1, or M1a1b that report Eastern European, or Southern European ancestry. I also have quite a few M33c, but I think they are Jewish matches. These are at 23andme. I have two matches over at FTDNA (smaller database) that are are M1a, but I don't remember the rest of the subclade as I'm getting the Uh oh Houston message now.

EDIT TO ADD-I almost forgot. I have a cousin who is just listed as M1a, and he is Arbëreshë. His family is from the same village as my family in Southern Italy. Though not Lemko, still interesting to me. :)

Jean M
09-21-2014, 08:54 PM
This would make it difficult to use mtDNA to distinguish Neolithic European farmers from later Indo European migrations if they have similar Neolithic mtDNA ancestry.

I've been saying this for some time. MtDNA has been really useful in distinguishing between Mesolithic and Neolithic population in Europe, but Y-DNA is the crucial difference between Neolithic and Indo-European populations. (Backed up by the full-genome work that has now distinguished between WHG, EEF and ANE.)

I suspect that we shall find, once we actually get enough samples from enough places and periods across Europe, that the Indo-European spread actually raised the level of mtDNA U5 and U4 from an extremely low level over most of Europe in the Neolithic to something closer to modern levels. Just having one U5/U4 per five people would do that. Razib Khan calls my theory "the revenge of the hunter-gatherers". :)

However I have a feeling that there may be one or two haplogroups that are distinctive to the Indo-European spread if we can refine by subclade, maybe subclades of H5a, T1a1'3 and U2e.

Generalissimo
09-22-2014, 02:31 AM
So the Yamnaya (and Proto-Indo Europeans) might be part of the Neolithic revolution, heavily influenced by pastoralists from the Near East, perhaps with a smaller influence from northern foragers from eastern Europe and central Asia.

You've got it sort of backwards there. The Proto-Indo-Europeans were probably a highly patriarchal and patrilineal society, but it's likely that they did incorporate lots of foreign women as they expanded. The results in Keyser et al. 2009 suggest this very clearly.

Link (http://hamagmongol.narod.ru/library/keyser_2009_e.pdf)

Also, there's no evidence of any expansion from the Near East into any part of Europe after the Neolithic, except places like Sicilly. I refer you to Lazaridis et al. 2014.

Link (http://arxiv.org/abs/1312.6639v2)

But perhaps you're confusing the Near East with the steppe and forest steppe of Eastern Europe?

parasar
09-22-2014, 03:15 AM
I've been saying this for some time. MtDNA has been really useful in distinguishing between Mesolithic and Neolithic population in Europe, but Y-DNA is the crucial difference between Neolithic and Indo-European populations. (Backed up by the full-genome work that has now distinguished between WHG, EEF and ANE.)

I suspect that we shall find, once we actually get enough samples from enough places and periods across Europe, that the Indo-European spread actually raised the level of mtDNA U5 and U4 from an extremely low level over most of Europe in the Neolithic to something closer to modern levels. Just having one U5/U4 per five people would do that. Razib Khan calls my theory "the revenge of the hunter-gatherers". :)

However I have a feeling that there may be one or two haplogroups that are distinctive to the Indo-European spread if we can refine by subclade, maybe subclades of H5a, T1a1'3 and U2e.

Two of the Motala samples are U2e1. N1a may be more of a distinguishing line.

nuadha
09-22-2014, 04:27 AM
You've got it sort of backwards there. The Proto-Indo-Europeans were probably a highly patriarchal and patrilineal society, but it's likely that they did incorporate lots of foreign women as they expanded. The results in Keyser et al. 2009 suggest this very clearly.

those weren't proto indo europeans. and of course the proto indo europeans spread their culture and language to different male lineages which is why the ydna of france, poland, greece, and latvia are all different.

BTW, the (highly likely) yamnaya migrations from the steppe to hungary were almost uniformly composed of children, men, and women.


Also, there's no evidence of any expansion from the Near East into any part of Europe after the Neolithic, except places like Sicilly. I refer you to Lazaridis et al. 2014.

So you dont really believe in your North Africa to Spain and beyond route for r1b in the post neolithic?

nuadha
09-22-2014, 04:30 AM
Two of the Motala samples are U2e1. N1a may be more of a distinguishing line.

agreed. it was also in mesolithic germany. using u2e to infer ie migrations is tenuous; not a complete waste but you probably need large numbers.

Generalissimo
09-22-2014, 04:38 AM
those weren't proto indo europeans. and of course the proto indo europeans spread their culture and language to different male lineages which is why the ydna of france, poland, greece, and latvia are all different.

Yes, but it took a while.


BTW, the (highly likely) yamnaya migrations from the steppe to hungary were almost uniformly composed of children, men, and women.

There were plenty of Indo-European male war bands running around Europe during the Bronze Age. They had to get their women somewhere.


So you dont really believe in your North Africa to Spain and beyond route for r1b in the post neolithic?

Bell Beakers expanded from Portugal. But they might have been partly of (secondary products) Neolithic North African origin.

GailT
09-22-2014, 06:50 AM
Also, there's no evidence of any expansion from the Near East into any part of Europe after the Neolithic, except places like Sicilly. I refer you to Lazaridis et al. 2014. But perhaps you're confusing the Near East with the steppe and forest steppe of Eastern Europe?

I'm referring to multiple waves of Neolithic migration from the Near East, broadly defined as the Middle, East and Anatolia, with at least two major waves of migration in the early Neolithic and mid/late Neolithic. The Indo-European migration from the Steppe/eastern Europe would be after the Neolithic farmers. Indo-European invaders brought their language so it seems likely they arrived as families and communities, not as bands of warriors who married local women.

The Keyser paper that you linked to is consistent with my statement that the Yamnaya could have been the product of Neolithic pastoralists from the Near East (mtDNA H, HV, T, K) expanding north and blending with northern foragers (mtDNA U5, U4, U2e).

Jean M
09-22-2014, 07:18 AM
Two of the Motala samples are U2e1.

Perhaps I wasn't clear. Here is what I said in AJ.


The rare haplogroup U2e seems Indo-European in its distribution, butappearances can deceive. It is far older than PIE. A recent estimate of its age was over 17,000 BC. ... We can deduce a presence on the European steppe at the time of PIE, as U2e appears in remains from archaeological cultures of west and east linked to the Indo-Europeans: Bell Beaker and Andronovo. It was also found in two Iron Age cemeteries, one near Qiemo, Xinjiang on the old southern Silk Road, and the other at Bøgebjerggård in Denmark. In modern populations U2e reaches a peak of nearly 16% among the Kalash of Northern Pakistan, a polytheistic people who speak an Indo-European language. East of that hot-spot there are lesser hot-spots along the former Silk Road, and to the west around the Black Sea. U2e seems to be scattered at lower levels right across the distribution of Indo-European languages.

In order to tell the difference between the U2e that spread with PIE (if any did) and other types of U2e, we would need to refine by subclade i.e. get deeper down the tree - not just U2e but subclades of U2e. Is this now a bit clearer?

Jean M
09-22-2014, 07:27 AM
N1a may be more of a distinguishing line.

Distinguishing of what? Surely you don't mean Indo-Europeans? N1a is so much more common in the European Neolithic than in Europe today that Haak saw it as the distinguishing lineage of the European Neolithic. http://www.ancestraljourneys.org/europeanneolithicdna.shtml

aarnisotka
09-22-2014, 07:38 AM
Also, there's no evidence of any expansion from the Near East into any part of Europe after the Neolithic

Then how to explain the Caucasus like ancestry in Europe. I've understood all the neolithic farmers are rather Basque/Sardinian or 'North African' like in ancestry.

Jean M
09-22-2014, 07:52 AM
using u2e to infer ie migrations is tenuous; not a complete waste but you probably need large numbers.

As I said above, mtDNA is not particularly useful in inferring IE migrations. I repeat for those who missed it the first time:


MtDNA has been really useful in distinguishing between Mesolithic and Neolithic population in Europe, but Y-DNA is the crucial difference between Neolithic and Indo-European populations. (Backed up by the full-genome work that has now distinguished between WHG, EEF and ANE.)


That case has now been clearly made, as I see it. (Of course some people will keep fighting the evidence until it bites back. So we need lots more data to join the dots.)

It is more of an academic interest whether we can track any mtDNA haplogroups along any specific IE routes. It would give us an idea of which movements were made by complete families, or the degree to which they were. But that is something that we can't do just at present, because so much of our aDNA data was gathered before the next-generation sequencing technologies made it possible to get full mtDNA genomes. It is not large numbers of samples that we need for this task so much as better quality results so that we can refine by subclade.

Jean M
09-22-2014, 08:06 AM
Then how to explain the Caucasus like ancestry in Europe. I've understood all the neolithic farmers are rather Basque/Sardinian or 'North African' like in ancestry.

What David meant by "after the Neolithic" was "later than the spread of Neolithic farmers into Europe".

In fact there is evidence of dairy farming developing later than the initial Neolithic and much closer to Europe - around the Sea of Marmara. As far as I can tell, piecing things together, some dairy farmers moved up the Danube and formed the first farming communities in the British Isles c. 4000 BC. Others moved east of the Carpathians to form the Cucuteni Culture, which interacted with Yamnaya.

Shaikorth
09-22-2014, 08:38 AM
Then how to explain the Caucasus like ancestry in Europe. I've understood all the neolithic farmers are rather Basque/Sardinian or 'North African' like in ancestry.

If it isn't of neolithic farmer origin - and it probably isn't if admixture calculators can be relied on this matter, Stuttgart for instance shows lots of Mediterranean and Basque type components but 0% West Asian in Eurogenes - that type of ancestry could have come from a place in Europe with no published samples as of yet. The PIE homeland like Samara valley is one. As for when the West Asian components might have gotten in there:



Samara experienced major population turnovers over time: early samples (>6000 years) belong primarily to mtDNA haplogroups U4 and U5, typical of European hunter-gatherers but later ones include haplogroups W, H, T, I, K, J.

So perhaps the West Asian/Caucasus component in Europe is from a group separate from Stuttgart & co that neolithised the Eastern European steppe before Indo-European expansion, or the product of its neolithisation.

Generalissimo
09-22-2014, 09:19 AM
Then how to explain the Caucasus like ancestry in Europe. I've understood all the neolithic farmers are rather Basque/Sardinian or 'North African' like in ancestry.

It depends what test you're talking about. Stuttgart, Oetzi and Gokhem2 do show Caucasus-specific components in many tests. And if they lack certain components that peak in the Caucasus and Central Asia, then this is probably because they lack ANE, rather than the fact that Europeans acquired more Near Eastern ancestry after the Neolithic. In other words, they lack ANE-specific alleles and Admixture is then forced to interpret that in the context of the clusters that it has to work with.

nuadha
09-22-2014, 09:39 AM
Perhaps I wasn't clear. Here is what I said in AJ.



In order to tell the difference between the U2e that spread with PIE (if any did) and other types of U2e, we would need to refine by subclade i.e. get deeper down the tree - not just U2e but subclades of U2e. Is this now a bit clearer?

its pretty clear you were talking about u2e specifically, and how it was typically found at significantly higher rates in IE people vs their non IE neighbors. after finding all the mesolithic u2e west of the steppe, im not sure your original sentiment is true.

as for going deeper down the tree, you could say that about any hg

aarnisotka
09-22-2014, 09:59 AM
It depends what test you're talking about. Stuttgart, Oetzi and Gokhem2 do show Caucasus-specific components in many tests. And if they lack certain components that peak in the Caucasus and Central Asia, then this is probably because they lack ANE, rather than the fact that Europeans acquired more Near Eastern ancestry after the Neolithic. In other words, they lack ANE-specific alleles and Admixture is then forced to interpret that in the context of the clusters that it has to work with.

Are you not now running in circles? Yes EEF samples do lack ANE, but it does not explain why Eastern Europeans score (in the latest MDLP fex) more of the caucasus specific component, where as Western Europeans and even Finns show more of a bias toward the EEF (Basque-like) component. Finns do not lack ANE, yet the program interprets Finns to have more of the EEF component. I guess you could speculate, that the Finnish ANE goes fex to the Siberian component, but yeah whatever.

On the oracle mode Finns also appear, at least in some cases, to be closer to the ancient farmer samples than fex Russians, yet the overall EEF ancestry in Russians is most definitely higher than in Finns.

Jean M
09-22-2014, 10:09 AM
its pretty clear you were talking about u2e specifically,

Here's what I said:


However I have a feeling that there may be one or two haplogroups that are distinctive to the Indo-European spread if we can refine by subclade, maybe subclades of H5a, T1a1'3 and U2e.

Only by ignoring all the words enlarged in red could anyone imagine that I was talking about U2e* itself as an IE marker. I do understand that the mind tends to skim. Readers on this forum will naturally focus on the haplogroups rather than the text in which they are embedded, so it is a lesson to me to try to be clearer in what I say. :biggrin1:

So let's start afresh. U2e* [the parent] cannot be an IE marker because it is Mesolithic [much earlier than Proto-Indo-European]. Some subclade(s) [daughter groups genetically, with extra mutations that differentiate them from the parent] of U2e might be IE markers. We have to wait and see. OK?

Generalissimo
09-22-2014, 10:15 AM
Are you not now running in circles? Yes EEF samples do lack ANE, but it does not explain why Eastern Europeans score (in the latest MDLP fex) more of the caucasus specific component, where as Western Europeans and even Finns show more of a bias toward the EEF (Basque-like) component. Finns do not lack ANE, yet the program interprets Finns to have more of the EEF component. I guess you could speculate, that the Finnish ANE goes fex to the Siberian component, but yeah whatever.

On the oracle mode Finns also appear, at least in some cases, to be closer to the ancient farmer samples than fex Russians, yet the overall EEF ancestry in Russians is most definitely higher than in Finns.

Obviously, you're the one running in circles because you're putting too much stock into ADMIXTURE results.

Most Europeans, except some groups, like Finns, can be modeled as mixtures of WHG, ANE and EEF. So all of the ADMIXTURE components that appear in most of Europe are combinations of these ancestral populations.

The fact that various ADMIXTURE clusters (like, for instance, those that peak among the Hazara, Ashkenazi, Kalash, etc.) appear among Europeans who can only be mixtures of WHG, ANE and EEF can only really mean we've looking at artifacts of the methodology.

Shaikorth
09-22-2014, 11:37 AM
The fact that various ADMIXTURE clusters (like, for instance, those that peak among the Hazara, Ashkenazi, Kalash, etc.) appear among Europeans who can only be mixtures of WHG, ANE and EEF can only really mean we've looking at artifacts of the methodology.

Not necessarily. The Hazara component's behaviour I've previously discussed, but the fact that the difference it makes between groups like CEU and North Russians is only quantitative (10% vs 25%) while it certainly is mainly ENA - 75% in Hazara and all their other components are more "western" - makes one wonder.

A successful 3-way fit using Stuttgart, Loschbour and MA-1 does not exclude more ancestral populations, just like 2-way fit being possible for some populations doesn't mean 3-way is not better, they mention this in the paper and point out two examples who already show discrepancy (Spanish and Estonians). The 3-way method isn't the best for detecting diluted ancestral populations either (modern Hungarians fit well).

The difference between Sardinian-centric and Caucasus-centric components is that while both have a Middle Eastern base, the other is WHG-shifted and the other ANE-shifted. The latter being spread by PIE's would make sense given its distribution in Europe north of the Mediterranean.
http://1.bp.blogspot.com/-LIug8kEPKW4/UFtC9XP-VsI/AAAAAAAAD1A/HTmsEscIGoI/s1600/MDLPwestasian.jpg

Iceman is 0% West Asian in this particular calculator. This doesn't mean Proto-Indo-European language came from the Near East, just that those who spread it had Basal Eurasian ancestry which maybe came to the far Eastern European steppe in the neolithic before the expansions (see this abstract (https://mcidublin.conference-services.net/reports/template/onetextabstract.xml?xsl=template/onetextabstract.xsl&conferenceID=3958&abstractID=814052)).

parasar
09-22-2014, 02:42 PM
Perhaps I wasn't clear. Here is what I said in AJ.



In order to tell the difference between the U2e that spread with PIE (if any did) and other types of U2e, we would need to refine by subclade i.e. get deeper down the tree - not just U2e but subclades of U2e. Is this now a bit clearer?

I did see that you mentioned subclades of U2e. The reason I noted the Motala U2e1s is because that particular subclade of U2e had been pointed to as a potential Indo-European marker before the Motala finds.
http://volgagermanbrit.us/documents/Kim_et_al.pdf

The genetic evidence of U2e1 and R1a1 may help to clarify the migration patterns of Indo-Europeans

parasar
09-22-2014, 02:55 PM
Distinguishing of what? Surely you don't mean Indo-Europeans? N1a is so much more common in the European Neolithic than in Europe today that Haak saw it as the distinguishing lineage of the European Neolithic. http://www.ancestraljourneys.org/europeanneolithicdna.shtml

N1a was meant to be seen as a negative inference - present in EEF (coming from its basal west eurasian) types, but absent in WHG and ANE types (before the ANE types admixed with EEF).

GailT
09-22-2014, 02:57 PM
n order to tell the difference between the U2e that spread with PIE (if any did) and other types of U2e, we would need to refine by subclade i.e. get deeper down the tree - not just U2e but subclades of U2e

Here is an example of the importance of having full sequence mtDNA samples with the most refined subclade possible. U5a1d2b is estimated to be about 6000 years old and has a typical modern Indo-European distrubtion. It's sister clade U5a1d2a is only found in in northwestern Europe, and the two share a common ancestor in U5a1d2 between 10,000 to 15,000 years ago. So having ancient samples from 4000 BC identified only as "U5a1d2" would not very helpful, since we really need to know if they are U5a1d2a or U5a1d2b.

This case is also an example of how only selected subclades of U5 reached the eastern Steppe in the Mesolithic. U5a1d2 probably originated in early central European hunter-gatherers, and only a subset, U5a1d2b, migrated to the Steppe and eventually was incorporated into early Steppe pastoralists (or at least this seems possible based on modern and ancient mtDNA). I described the large variation in population of U5 subclades on the FTDNA project page, and most of U5 in Europe today is in a few subclades that expanded in size rapidly in the Neolithic, perhaps either in early farmers or in Steppe peoples. This is consistent with Jean's theory of the "revenge" of the hunter-gatherers. But we really need full mtDNA sequence samples to test this theory.

I still think there is potential for the use of mtDNA, when we have more modern and ancient samples, to help in the analysis of fine scale populations movements.

Jean M
09-22-2014, 03:54 PM
I still think there is potential for the use of mtDNA, when we have more modern and ancient samples, to help in the analysis of fine scale populations movements.

I certainly hope so Gail. I really appreciate the work you have been doing.

nuadha
09-22-2014, 05:11 PM
Here's what I said:



Only by ignoring all the words enlarged in red could anyone imagine that I was talking about U2e* itself as an IE marker. I do understand that the mind tends to skim. Readers on this forum will naturally focus on the haplogroups rather than the text in which they are embedded, so it is a lesson to me to try to be clearer in what I say. :biggrin1:

I was responding to the quote you posted from AJ. There you do say that u2e is a marker for PIE. You say so because of the elevated levels of u2e in IE people relative to non IE, which I now doubt is the case. Clearly you were not saying that u2e is only found in IE but you did say its found at higher rates which rightfully would make it a marker if true.

The rare haplogroup U2e seems Indo-European in its distribution, butappearances can deceive. It is far older than PIE. A recent estimate of its age was over 17,000 BC. ... We can deduce a presence on the European steppe at the time of PIE, as U2e appears in remains from archaeological cultures of west and east linked to the Indo-Europeans: Bell Beaker and Andronovo. It was also found in two Iron Age CEMETERIES, one near Qiemo, Xinjiang on the old southern Silk Road, and the other at Bøgebjerggård in Denmark. In modern populations U2e reaches a peak of nearly 16% among the Kalash of Northern Pakistan, a polytheistic people who speak an Indo-European language. East of that hot-spot there are lesser hot-spots along the former Silk Road, and to the west around the Black Sea. U2e seems to be scattered at lower levels right across the distribution of Indo-European languages.

Having said that I would rather have an actual discussion on whether u2e was elevated in IE peoples rather than figuring out what jean said.


So let's start afresh. U2e* [the parent] cannot be an IE marker because it is Mesolithic [much earlier than Proto-Indo-European]. Some subclade(s) [daughter groups genetically, with extra mutations that differentiate them from the parent] of U2e might be IE markers. We have to wait and see. OK?

cant your say that about any haplogroup? H, K, I, J, ...

Jean M
09-22-2014, 06:13 PM
You say so because of the elevated levels of u2e in IE people relative to non IE.

Heavens to Betsy, Nuada! I was having a first approximation at a guess based on the only data I had, which was for U2e in toto. That distribution might be totally misleading. Or it might turn out to mainly belong to one subclade which happens to be one found in Yamnaya. Or whatever. We cannot at this moment sort this out.


cant your say that about any haplogroup? H, K, I, J, ...

Yes exactly. We are now on the same page. Specific subclades are what to look for. Not elevated levels of the parent clade.

alan
09-22-2014, 06:23 PM
Heavens to Betsy, Nuada! I was having a first approximation at a guess based on the only data I had, which was for U2e in toto. That distribution might be totally misleading. Or it might turn out to mainly belong to one subclade which happens to be one found in Yamnaya. Or whatever. We cannot at this moment sort this out.



Yes exactly. We are now on the same page. Specific subclades are what to look for.

I like Heavens to Betsy - are those the words you actually used :biggrin1:

Jean M
09-22-2014, 06:27 PM
I like Heavens to Betsy - are those the words you actually used

Yep. I'm very ladylike.

Hando
09-22-2014, 07:50 PM
I suspect that we shall find, once we actually get enough samples from enough places and periods across Europe, that the Indo-European spread actually raised the level of mtDNA U5 and U4 from an extremely low level over most of Europe in the Neolithic to something closer to modern levels. Just having one U5/U4 per five people would do that. Razib Khan calls my theory "the revenge of the hunter-gatherers". :)
If IE mtDNA was more Neolithic than Mesolithic in character how could IE raise the level of U5/U4 in Europeans? Aren't U5/U4 Mesolithic hunter gatherer DNA?

Hando
09-22-2014, 07:56 PM
You've got it sort of backwards there. The Proto-Indo-Europeans were probably a highly patriarchal and patrilineal society, but it's likely that they did incorporate lots of foreign women as they expanded. The results in Keyser et al. 2009 suggest this very clearly.
But perhaps you're confusing the Near East with the steppe and forest steppe of Eastern Europe?
Sorry, but what does GailT have backwards?
She said "So the Yamnaya (and Proto-Indo Europeans) might be part of the Neolithic revolution, heavily influenced by pastoralists from the Near East, perhaps with a smaller influence from northern foragers from eastern Europe and central Asia."

Jean M
09-22-2014, 08:25 PM
If IE mtDNA was more Neolithic than Mesolithic in character how could IE raise the level of U5/U4 in Europeans?

As I said earlier Hando, this particular study of a handful of individuals is not going to tell us the real, genuine, actual, reliable facts about the relative proportions of various mtDNA haplogroups across the entire Yamnaya Horizon. Perhaps I should have been more explicit. I don't believe the 4 or 5 results tell us the true story. I don't believe that there was no U5 or U4 in Yamnaya.

GailT
09-22-2014, 09:42 PM
If IE mtDNA was more Neolithic than Mesolithic in character how could IE raise the level of U5/U4 in Europeans? Aren't U5/U4 Mesolithic hunter gatherer DNA?

If you look at the European Neolithic ancient mtDNA studies, there is very little U5 or U4, even less than the present day population in Europe. One possibility is that Indo European speakers had a larger percentage of U5 and U4 than did Neolithic farmers, so their arrival in central/western Europe increased the percentage of U5 & U4. Another possibility is that U5 and U4 were present at the fringe of northern and western and merged with farmers as they expanded into the fringe. Some combination of the two might explain why U5 & U4 increased in Europe after the Neolithic.

Hando
09-23-2014, 02:36 AM
If you look at the European Neolithic ancient mtDNA studies, there is very little U5 or U4, even less than the present day population in Europe. One possibility is that Indo European speakers had a larger percentage of U5 and U4 than did Neolithic farmers, so their arrival in central/western Europe increased the percentage of U5 & U4. Another possibility is that U5 and U4 were present at the fringe of northern and western and merged with farmers as they expanded into the fringe. Some combination of the two might explain why U5 & U4 increased in Europe after the Neolithic.

And this mixed group from the northern and western fringes then proceeded to move into more central areas? If so, are there any answers to as who, when and why? I ask this because if this mixed group did indeed migrate into more central areas with their U4 and U5, then it would appear that they were strong enough to encroach on land occupied by EEF groups and multiply enough so that modern Europeans have enough of their genes. Who would be strong enough to encroach onto land occupied by technologically advanced EEF?

Generalissimo
09-23-2014, 02:43 AM
Who would be strong enough to encroach onto land occupied by technologically advanced EEF?

Probably someone who became even more technologically and socially advanced.

A biocultural perspective on the transition to agriculture in Central Europe (http://puvodni.mzm.cz/Anthropologie/article.php?ID=1549)

GailT
09-23-2014, 04:10 AM
And this mixed group from the northern and western fringes then proceeded to move into more central areas? If so, are there any answers to as who, when and why? I ask this because if this mixed group did indeed migrate into more central areas with their U4 and U5, then it would appear that they were strong enough to encroach on land occupied by EEF groups and multiply enough so that modern Europeans have enough of their genes.

I expect it was more a blending of hunter-gatherer and farming communities. For example, the Bollongino studies shows separate populations co-existing. Percentages of U5 increase as you go north, with 8.5% U5 in Danes (see the Li et al U5 samples, link (http://www.anthrogenica.com/showthread.php?1417-New-U5-sequences)) compared to about 5% in southern Europe. It's interesting that about half of the Li U5 samples are in 4 subclades (U5a1a, U5a1b, U5a2a and U5a2b) and all four expanded rapidly in the Neolithic. Perhaps these 4 subclades returned to central.western Europe with Indo-European migrations from the Steppe? Or perhaps as early migration that brought ANE to northern Europe?

For comparison of U5 diversity in western Europe, percentages of U5 subclades in Asturians and Basques are in the jpeg.

Hando
09-23-2014, 07:48 PM
I expect it was more a blending of hunter-gatherer and farming communities. For example, the Bollongino studies shows separate populations co-existing. Percentages of U5 increase as you go north, with 8.5% U5 in Danes (see the Li et al U5 samples, link (http://www.anthrogenica.com/showthread.php?1417-New-U5-sequences)) compared to about 5% in southern Europe. It's interesting that about half of the Li U5 samples are in 4 subclades (U5a1a, U5a1b, U5a2a and U5a2b) and all four expanded rapidly in the Neolithic. Perhaps these 4 subclades returned to central.western Europe with Indo-European migrations from the Steppe? Or perhaps as early migration that brought ANE to northern Europe?

For comparison of U5 diversity in western Europe, percentages of U5 subclades in Asturians and Basques are in the jpeg.
So you are suggesting a migration of U5 from the Steppes back into central and Western Europe. But I was wondering whether the reemergence of U5 was due to mixed or blended as you say populations from the northern and Western fringes. Do you not agree then?

evon
09-23-2014, 08:26 PM
So you are suggesting a migration of U5 from the Steppes back into central and Western Europe. But I was wondering whether the reemergence of U5 was due to mixed or blended as you say populations from the northern and Western fringes. Do you not agree then?

Ive noticed that ive got three FGS matches in Europe, in Western Sweden and Krakow Poland. I wonder if this is due to a northward movement of my U5 linage via Central/Eastern Europe into Scandinavia..

GailT
09-24-2014, 12:23 AM
So you are suggesting a migration of U5 from the Steppes back into central and Western Europe. But I was wondering whether the reemergence of U5 was due to mixed or blended as you say populations from the northern and Western fringes. Do you not agree then?

My guess is that it was a combination both migration from the steppe and blending with with U5 hunter-gatherers at the fringe.

newtoboard
03-01-2015, 04:29 PM
It is unfortunate we won't see this paper until 2017. I am still confused what Catacomb represents. Catcaomb like characteristics seems to first appear in the Northern Don region, in the Southern Dnieper region and Corded Ware. If the Indo-Iranian split into Iranian and Indo-Aryan is closer to the 2800-2500 BC hypothesis than the 2000-1800 BC hypothesis would this mean there would have to be at least two cultures around that time representing this split? Something like

Poltavka, Potpovka-Indo-Aryan(+Dardic)+Nuristani
Catacomb, Abashevo =Iranian+Scythio-Siberian+Cimmerian

Does that sound reasonable? I would expect the Iranian related cultures to have higher WHG components which seems to be true today as well. And they would be placed in a region where we could account for Iranian related hydronyms (Dniester, Dnieper, Don etc). Or could that split have occurred within the eastern cultures (Poltavka, Potpovka)? As far as I know those cultures don't have different variants like Andronovo does with Alakul and Federovo to account for at least two different linguistic groups.

DMXX
03-03-2015, 09:26 PM
Here's an idea which came up while discussing this with a fellow enthusiast recently.

The Samara Yamnaya samples (~3300-2700 B.C.) were modelled as half EHG (peaking in the two HG's) and half of the West Asian-like teal component in the original runs. The latter component clearly made an appearance in the region during the Neolithic.

What if gene flow from wherever the progenitor region was continued into later steppe groups?

The Poltavka culture (http://en.wikipedia.org/wiki/Poltavka_culture) (descendant of Yamnaya, ancestor of Sintashta) is commonly linked to the earliest proto-Indo-Iranian speakers. The dating of this culture is essentially in-between Yamnaya and Andronovo (2700-2100 B.C.). On top of this, Poltavka is situated in roughly the same region of the steppes as these Samara Yamnaya samples.

If gene flow continued from the progenitor source of the West Asian-like component (Wikipedia states there allegedly was Maikop influence), one would expect the early proto-Indo-Iranians to contain more of this component and less EHG. This does fit with the precedence set between the Samara HG and Yamnaya samples (increase of the West Asian-like component with time).

As such, modelling the extent of Samara Yamnaya admixture in Asian populations may present with a misleading picture if the intention is to quantify the Bronze age steppe contribution on a genetic level, with whatever values produced inherently underestimating the actual extent. There is no reason to assume complete genetic congruence between the Samara Yamnaya samples and Andronovo (seven century juncture).

alan
03-05-2015, 12:41 PM
Eurogenes
t[I]he best two-way admixture model for the Yamnaya genomes, from the Samara region near the Russo-Kazakh border, is Samara_HG/Tabassaran, rather than Karelia_HG/Armenian as per Haak et al., which appears to be a better solutio

Tabassaran is a Lezgic NE Caucasian language in Dagestan on the Caspian shore. What strikes me about this is Maykop not only was in the Kuban sort of area but stretched down to NW Iran where Maykop Kurgans were located. Dagestan lies on that axis. We know Maykop started to be influential and also have small pockets of settlement on the steppe from 3500BC in the last couple of centuries before Yamnaya and some argue that many things in Yamnaya are prefigured in Maykop. Even if settlement was very modest and late on the steppes, it is still tempting to think that gene flow - even if mainly intermarriage of steppe men and Maykop women for alliance and trade - could have been significant at least in the elites, and it is the elites we are seeing in those Kurgans.

I am not saying that all Caucasian influence started only with Maykop. I have seen it argued that pre-Maykop farmer influence can also be seen in parts of the steppe too. It may have been a continuum.

Also, and this might be crucial to this arguement, the last paper of Maykop-the one in German-emphasised that the Mesopotamian links dont stand up to scrutiny and instead the main source of influence in Maykop seems to have been northern Iran and even Turkmenistan. So, given that an inner asian/west central Asian element has been suggested in the non-HG part of Yamnaya, this fits rather well too.

An upshot of this idea would of course be that if Suvorovo/Sredny Stog offshoot=Anatolian is IE then given its age c. 4300BC then it would have not had Maykop influence and genes. This would kind of confirm that the Maykop influence may have been important in terms of the development of Yamnaya but it was not the bringer of IE in its earliest forms. It was the local HG population who carried the core root of IE. This of course can be tested if Sredny Stog and other pre-Yamnaya steppe and steppe-offshoot groups are tested and found to lack some of the non-HG/Caucasian input that is found in Yamnaya.

The issue of the pre-Yamnaya Anatolian type PIE branch and what y lines would represent it if Anthony's ideas are correct is interesting but I am going to start another thread on this.

newtoboard
03-05-2015, 12:55 PM
Eurogenes
t[I]he best two-way admixture model for the Yamnaya genomes, from the Samara region near the Russo-Kazakh border, is Samara_HG/Tabassaran, rather than Karelia_HG/Armenian as per Haak et al., which appears to be a better solutio

Tabassaran is a Lezgic NE Caucasian language in Dagestan on the Caspian shore. What strikes me about this is Maykop not only was in the Kuban sort of area but stretched down to NW Iran where Maykop Kurgans were located. Dagestan lies on that axis. We know Maykop started to be influential and also have small pockets of settlement on the steppe from 3500BC in the last couple of centuries before Yamnaya and some argue that many things in Yamnaya are prefigured in Maykop. Even if settlement was very modest and late on the steppes, it is still tempting to think that gene flow - even if mainly intermarriage of steppe men and Maykop women for alliance and trade - could have been significant at least in the elites, and it is the elites we are seeing in those Kurgans.

I am not saying that all Caucasian influence started only with Maykop. I have seen it argued that pre-Maykop farmer influence can also be seen in parts of the steppe too. It may have been a continuum.

Also, and this might be crucial to this arguement, the last paper of Maykop-the one in German-emphasised that the Mesopotamian links dont stand up to scrutiny and instead the main source of influence in Maykop seems to have been northern Iran and even Turkmenistan. So, given that an inner asian/west central Asian element has been suggested in the non-HG part of Yamnaya, this fits rather well too.

An upshot of this idea would of course be that if Suvorovo/Sredny Stog offshoot=Anatolian is IE then given its age c. 4300BC then it would have not had Maykop influence and genes. This would kind of confirm that the Maykop influence may have been important in terms of the development of Yamnaya but it was not the bringer of IE in its earliest forms. It was the local HG population who carried the core root of IE. This of course can be tested if Sredny Stog and other pre-Yamnaya steppe and steppe-offshoot groups are tested and found to lack some of the non-HG/Caucasian input that is found in Yamnaya.

The issue of the pre-Yamnaya Anatolian type PIE branch and what y lines would represent it if Anthony's ideas are correct is interesting but I am going to start another thread on this.

Is there really any option other than Z2103?

Dr_McNinja
03-05-2015, 02:25 PM
I would expect the Iranian related cultures to have higher WHG components which seems to be true today as well.Correct me if I'm wrong, I think WHG is higher in Central Asia (Tajiks, Turkmen), South Central (Afghans), and parts of South Asia (Jatts, North Indian Brahmins) than Iran. In the K8 results I've seen 5.41% WHG in a Kandahari Pashtun (and overall a higher average among Afghans than Iranians) and 7.41% WHG in a Jatt from Haryana, North India (HRP0393 who was lower than a few other Haryana individuals in WHG in ANE K7, so I think we'd see as high as 8% in many of them).

I think we need ancient Iranian DNA to be sure. Iran might have been heavily influenced and had older WHG levels diluted by subsequent mixing with other Near Eastern groups in the last two thousand years, especially post-Islam.

newtoboard
03-05-2015, 02:44 PM
Correct me if I'm wrong, I think WHG is higher in Central Asia (Tajiks, Turkmen), South Central (Afghans), and parts of South Asia (Jatts, North Indian Brahmins) than Iran. In the K8 results I've seen 5.41% WHG in a Kandahari Pashtun (and overall a higher average among Afghans than Iranians) and 7.41% WHG in a Jatt from Haryana, North India (HRP0393 who was lower than a few other Haryana individuals in WHG in ANE K7, so I think we'd see as high as 8% in many of them).

I think we need ancient Iranian DNA to be sure. Iran might have been heavily influenced and had older WHG levels diluted by subsequent mixing with other Near Eastern groups in the last two thousand years, especially post-Islam.

I meant Iranian as in those who spoke Iranian languages so that includes Pamiris, Pashtuns who do seem to have some WHG. Jatts have some WHG but you can't honestly not find something suspicious about the group which is said to have Scythian related ancestry having high WHG for South Asia. Or about the fact that WHG is not on any sort of continuum in South Asia. It is extrmely uneven and I think some North Indian groups will have close to 0 and certainly less than 5%. The uneveness of it is what makes me think it arrived later with Iranian speaking groups like Scythians, Sakas, Indo-Parthians, and Kushans etc

DMXX
03-05-2015, 03:07 PM
Correct me if I'm wrong, I think WHG is higher in Central Asia (Tajiks, Turkmen), South Central (Afghans), and parts of South Asia (Jatts, North Indian Brahmins) than Iran. In the K8 results I've seen 5.41% WHG in a Kandahari Pashtun (and overall a higher average among Afghans than Iranians) and 7.41% WHG in a Jatt from Haryana, North India (HRP0393 who was lower than a few other Haryana individuals in WHG in ANE K7, so I think we'd see as high as 8% in many of them).


This is correct.



I think we need ancient Iranian DNA to be sure. Iran might have been heavily influenced and had older WHG levels diluted by subsequent mixing with other Near Eastern groups in the last two thousand years, especially post-Islam.

Davidski made a similar assertion recently with respect to ANE frequencies in the region. That does not appear substantiated given that Kurds and Iranians sit in a predictable location on an ANE frequency continuum running from SE Turkey through to Tajikistan. If there was substantial post-Islamic mixing in the region (anyone familiar with the history of the region is aware this is an improbable proposition anyway), we would expect a depreciation in ANE, given Southwest Asia evidently has less ANE than other parts of Eurasia. That isn't what the data suggests (http://www.anthrogenica.com/showthread.php?30-Eurogenes-Project-thread&p=70483#post70483).

Another piece of evidence which dissuades me away from any significant reshuffling of components in Iran following the Islamic expansion are the SW Asian component frequencies. For instance, in Dodecad K12b (https://docs.google.com/spreadsheet/ccc?key=0ArJDEoCgzRKedEY4Y3lTUVBaaFp0bC1zZlBDcTZEY lE&hl=en_US#gid=0), the Iranian population averages score 12.4 (Iranian_D) and 14.2% (Iranians from Behar et al.*). The Armenian average across the three populations (_D, Behar, Yunusbayev) is 12.5% (ranges from 9.9-14.3%). Thus, despite being situated north of Iranians geographically, maintaining endogamy since ancient times and being Christian South Caucasians, the Armenians have an equivalent value. This too makes the notion of component reshuffling across the region after Islam's expansion untenable.

Summarised, the totality of the autosomal evidence has suggested to me that, rather than Iranians (and Kurds) becoming more "Near-Eastern" over time, their genetic architecture was mostly crystallised at a more distant point in time than 600 A.D.; it looks to me as if the only significant genetic contribution brought to West Asia since 0 A.D. were the Oghuz Turks.

Aside from the scenario where an Andronovo-mediated steppe contribution happens to be stronger in Central and South Asians than Iranians, another possibility is that some WHG (and ANE?) filtered down towards South-Central Asia before the Indo-European expansions into South-Central Asia took place. Molodin et al. sampled hunter-gatherer specimens from Central Eurasia. I blogged about these results shortly afterwards (http://vaedhya.blogspot.co.uk/2012/07/secrets-of-central-asia-chapter-ii.html) (please look for the Krotovo-Petrovo culture interactions).

* As demonstrated by Zack at the Harappa project (http://www.harappadna.org/2011/03/iranians/), three of the Behar samples - GSM536746, GSM536751 and GSM536758 - score considerably more SW Asian than the other Iranians (sudden 5% increase between GSM536750 and GSM536746). According to Baibars, Behar's samples were taken from across Iran. Also note these three also have significant (>5%) Sub-Saharan African ancestry. This would suggest they are from the deep south of the country. As such, recent Arab ancestry may be a possible explanation for the increased SW Asian. If these individuals were excluded, the Behar SW Asian average will probably be comparable to the 12.4% range produced by Dienekes.

newtoboard
03-05-2015, 03:43 PM
This is correct.



Davidski made a similar assertion recently with respect to ANE frequencies in the region. That does not appear substantiated given that Kurds and Iranians sit in a predictable location on an ANE frequency continuum running from SE Turkey through to Tajikistan. If there was substantial post-Islamic mixing in the region (anyone familiar with the history of the region is aware this is an improbable proposition anyway), we would expect a depreciation in ANE, given Southwest Asia evidently has less ANE than other parts of Eurasia. That isn't what the data suggests (http://www.anthrogenica.com/showthread.php?30-Eurogenes-Project-thread&p=70483#post70483).

Another piece of evidence which dissuades me away from any significant reshuffling of components in Iran following the Islamic expansion are the SW Asian component frequencies. For instance, in Dodecad K12b (https://docs.google.com/spreadsheet/ccc?key=0ArJDEoCgzRKedEY4Y3lTUVBaaFp0bC1zZlBDcTZEY lE&hl=en_US#gid=0), the Iranian population averages score 12.4 (Iranian_D) and 14.2% (Iranians from Behar et al.*). The Armenian average across the three populations (_D, Behar, Yunusbayev) is 12.5% (ranges from 9.9-14.3%). Thus, despite being situated north of Iranians geographically, maintaining endogamy since ancient times and being Christian South Caucasians, the Armenians have an equivalent value. This too makes the notion of component reshuffling across the region after Islam's expansion untenable.

Summarised, the totality of the autosomal evidence has suggested to me that, rather than Iranians (and Kurds) becoming more "Near-Eastern" over time, their genetic architecture was mostly crystallised at a more distant point in time than 600 A.D.; it looks to me as if the only significant genetic contribution brought to West Asia since 0 A.D. were the Oghuz Turks.

Aside from the scenario where an Andronovo-mediated steppe contribution happens to be stronger in Central and South Asians than Iranians, another possibility is that some WHG (and ANE?) filtered down towards South-Central Asia before the Indo-European expansions into South-Central Asia took place. Molodin et al. sampled hunter-gatherer specimens from Central Eurasia. I blogged about these results shortly afterwards (http://vaedhya.blogspot.co.uk/2012/07/secrets-of-central-asia-chapter-ii.html) (please look for the Krotovo-Petrovo culture interactions).

* As demonstrated by Zack at the Harappa project (http://www.harappadna.org/2011/03/iranians/), three of the Behar samples - GSM536746, GSM536751 and GSM536758 - score considerably more SW Asian than the other Iranians (sudden 5% increase between GSM536750 and GSM536746). According to Baibars, Behar's samples were taken from across Iran. Also note these three also have significant (>5%) Sub-Saharan African ancestry. This would suggest they are from the deep south of the country. As such, recent Arab ancestry may be a possible explanation for the increased SW Asian. If these individuals were excluded, the Behar SW Asian average will probably be comparable to the 12.4% range produced by Dienekes.

Wouldn't that apply to Iran too in the presence of M343, P25, M269 and Z2103? It almost certainly applies for South Asia since R2 is significant. But one thing that jumped out to me (and a chart of R1+R2 frequency vs ANE would verify this) is that ANE seems to be at a very similar level across South Asia and South Central Asia (referring mostly to Afghanistan and Tajikistan) regardless of whether the group is mostly R1a rich, R2 rich or has a balanced distribution of the two. To me this suggests that the natives of these regions as well as the incoming Indo-Iranian migrants had a similar level of ANE. So where R1 is heavy most of the ANE is from IE migrants, where R2 is rich most will be pre-Indo-European and the mixed regions will obviously have a significant amount of both. So pre IE ANE could have been replaced by IE ANE without pre IE ANE inflating ANE in these regions.

Dr_McNinja
03-05-2015, 04:38 PM
I meant Iranian as in those who spoke Iranian languages so that includes Pamiris, Pashtuns who do seem to have some WHG. Jatts have some WHG but you can't honestly not find something suspicious about the group which is said to have Scythian related ancestry having high WHG for South Asia. Or about the fact that WHG is not on any sort of continuum in South Asia. It is extrmely uneven and I think some North Indian groups will have close to 0 and certainly less than 5%. The uneveness of it is what makes me think it arrived later with Iranian speaking groups like Scythians, Sakas, Indo-Parthians, and Kushans etcThere is a continuum of sorts. It peaks in Haryana as a sort of ground zero, then drops around Nepal Brahmins, then North Indian Brahmins on the eastern side and on the western side sharply drops among Jatts in Punjab as the component seems to compete with Gedrosian. It's pretty easy to predict a person's area of ancestry going by their WHG/European admixture levels relative to Haryana as ground zero. The peak is kept better among Nepalese Brahmins I think because other North Indian Brahmins (like from Uttar Pradesh) migrated around more. Rajasthani Jatts, the 2 or 3 we've seen in Harappa Project, were very close to Haryanvi individuals.

Then it hits another peak in Southeastern Afghanistan (Kandahar). From there the gradient goes northwards to Tajiks, but it drops sharply among Pashtun to the east because of, again, the Gedrosian component competing with it.

Eurogenes ANE K7 https://docs.google.com/spreadsheet/ccc?key=0AuXBmvmgdkfVdFMtRHVlZDBuQ3lMcjhxMDE4V3JoY lE#gid=22

West Eurasia K8 https://docs.google.com/spreadsheet/ccc?key=0AuXBmvmgdkfVdFMtRHVlZDBuQ3lMcjhxMDE4V3JoY lE#gid=35

Visualizing this on a map, this gradient seems like a stereotypical depiction of the Indo-Aryan spread, even the Afghan part, albeit interrupted by Gedrosian.

Another competitor component is East Eurasian/Oceanian which is elevated near the Himalayas, more noticeable among Western Pahari groups (Kashmir) since they've got both Gedrosian and this East Eurasian/Oceanian combo. The Nepalese don't have the Gedrosian and the Central/Eastern Punjabi Jatts don't have East Eurasian/Oceanian, so they're the next highest in WHG.

I have no idea why Oceanian is raised near the mountainous areas and doesn't correlate straight with ASE except in actual South Indians.

Dr_McNinja
03-05-2015, 04:49 PM
Summarised, the totality of the autosomal evidence has suggested to me that, rather than Iranians (and Kurds) becoming more "Near-Eastern" over time, their genetic architecture was mostly crystallised at a more distant point in time than 600 A.D.; it looks to me as if the only significant genetic contribution brought to West Asia since 0 A.D. were the Oghuz Turks. The impact of those Turkic groups also makes me wonder, but they'd be high in WHG too I figure, going by current Central Asian groups. Hard to say though because the Turkic tribes vary greatly in autosomal admixture depending on where and when we're talking about.

In any case, all I meant to say that was without ancient Iranian DNA giving a surprising amount of WHG/ANE, we wouldn't be able to form a coherent picture linking Iran as a possible source for the elevated WHG among some groups in South Central Asia. Right now Iran is right where it should be in the background gradient.


Aside from the scenario where an Andronovo-mediated steppe contribution happens to be stronger in Central and South Asians than Iranians, another possibility is that some WHG (and ANE?) filtered down towards South-Central Asia before the Indo-European expansions into South-Central Asia took place. Molodin et al. sampled hunter-gatherer specimens from Central Eurasia. I blogged about these results shortly afterwards (http://vaedhya.blogspot.co.uk/2012/07/secrets-of-central-asia-chapter-ii.html) (please look for the Krotovo-Petrovo culture interactions). I like this scenario myself.

DMXX
03-05-2015, 05:01 PM
The impact of those Turkic groups also makes me wonder, but they'd be high in WHG too I figure, going by current Central Asian groups. Hard to say though because the Turkic tribes vary greatly in autosomal admixture depending on where and when we're talking about.


One possibility here is the majority of WHG currently found in Central Asia was already present before the Turkish migrations began. Localised demic events (de+repopulation, migrations, genetic drift, resettlements from Soviet times, you name it) would have no doubt blurred the picture further.



In any case, all I meant to say that was without ancient Iranian DNA giving a surprising amount of WHG/ANE, we wouldn't be able to form a coherent picture linking Iran as a possible source for the elevated WHG among some groups in South Central Asia. Right now Iran is right where it should be in the background gradient.


Agreed. In an Iranian (the country) context, some samples from the Yaz culture (http://en.wikipedia.org/wiki/Yaz_culture) would prove useful. Wiki claims it's associated with East Iranic languages, but at least one paper I read some years back linked it with proto-West-Iranic speakers. Apparently no burials have been found though.

Jean M
03-05-2015, 06:32 PM
It seems likely that the actual frequencies of farmer mtDNA varied across the steppe, with higher levels in the centre and west than the east.

Following up this earlier discussion: we now have 34 Yamnaya samples yielding mtDNA: (Details with locations: 3961 )

Mesolithic type mtDNA (total 8):
U
U2e1a
U4a1
U5a1a1 (2 samples)
U5
U5a1 (2 samples)

Neolithic type mtDNA (total 26)
H (7 samples)
H2b
H6a1b
H13a1a1a
I
J
K (2 samples)
N1a
T (2 samples)
T1
T1a (2 samples)
T2c1a2
T2a1b1a
W
W3a1a
W6
X

parasar
03-05-2015, 07:32 PM
Following up this earlier discussion: we now have 34 Yamnaya samples yielding mtDNA: (Details with locations: 3961 )

Mesolithic type mtDNA (total 8):
U
U2e1a
U4a1
U5a1a1 (2 samples)
U5
U5a1 (2 samples)

Neolithic type mtDNA (total 26)
H (7 samples)
H2b
H6a1b
H13a1a1a
I
J
K (2 samples)
N1a
T (2 samples)
T1
T1a (2 samples)
T2c1a2
T2a1b1a
W
W3a1a
W6
X

Interestingly not one mt-M line.
Do you suppose the M lines came in with R1a?
"three Neolithic and all three Bronze Age individuals belonged to mtDNA haplogroup C, which is common in East Eurasian, particularly South Siberian, populations but exceedingly rare in Europe ... Ni58, Ya45, and Ya34 are members of haplogroup C, which is one of the most common mtDNA lineages in East Eurasia and likely originated in South Siberia ... All three Bronze Age samples (D1.8, L8, and L15) also belong to haplogroup C, bringing the total number of haplogroup C individuals in the present study to six ... thought to have originated in South Siberia in the area surrounding Lake Baikal "
http://scholarworks.gvsu.edu/cgi/viewcontent.cgi?article=1004&context=theses

Table 1
https://digital.library.adelaide.edu.au/dspace/bitstream/2440/74221/1/02whole.pdf
https://digital.library.adelaide.edu.au/dspace/bitstream/2440/74221/2/01front.pdf

parasar
03-05-2015, 08:54 PM
Following up this earlier discussion: we now have 34 Yamnaya samples yielding mtDNA: (Details with locations: 3961 )

Mesolithic type mtDNA (total 8):
U
U2e1a
U4a1
U5a1a1 (2 samples)
U5
U5a1 (2 samples)

Neolithic type mtDNA (total 26)
H (7 samples)
H2b
H6a1b
H13a1a1a
I
J
K (2 samples)
N1a
T (2 samples)
T1
T1a (2 samples)
T2c1a2
T2a1b1a
W
W3a1a
W6
X

And then, the mt-N types went west with R1a1.
Hard to believe it is 7.5 years since ryschev's post.
http://forums.familytreedna.com/showthread.php?t=945&page=4

Anterior 06-sep-2007, 02:21
rsychev rsychev est
Registered User

Registrado: julio-2006
Ubicaci: Moskovia
Posts: 178
STR typing of ancient DNA extracted from hair shafts of Siberian mummies.
Amory S, Keyser C, Crube'zy E, Ludes B.

Andronovo culture (2300?1000 BC)
http://en.wikipedia.org/wiki/Andronovo_culture

yDNA - R1a1
16-11,14-14-32-25-11-11-13-14-11-10
(DYS19-385a,b-389I-389II-390-391-392-393-437-438-439) - 2 sample

mtDNA - T1, U4, U2e, H?, K, U5a1

Tagar culture (800?200 BC)
http://en.wikipedia.org/wiki/Tagar_culture

yDNA - R1a1
17-11,14-13-31-24-11-11-13-14-11-10
16-11,14-13-31-24-11-11-13-14-11-10
16-11,14-14-31-25-11-11-13-14-11-10
17-11,14-13-31-24-11-12-13-14-11-10
(DYS19-385a,b-389I-389II-390-391-392-393-437-438-439)

Tachtyk culture (200 BC?200 AD)
http://en.wikipedia.org/wiki/Tashtyk_culture

yDNA - R1a1
17-11,14-13-31-24-11-11-13-14-11-10
(DYS19-385a,b-389I-389II-390-391-392-393-437-438-439)

Afanassievo culture (3500?2500 BC)
yDNA - no data
mtDNA - N1a, T

Jean M
03-05-2015, 09:33 PM
Interestingly not one mt-M line. Do you suppose the M lines came in with R1a?

MtDNA M is not bulking large in my thinking about the Yamnaya or its descendants, because it does not occur in any of the tested samples from Yamnaya, Andronovo, Catacomb, Corded War or Bell Beaker, and there is just one M from the Tarim Basin mummies.

parasar
03-05-2015, 10:16 PM
MtDNA M is not bulking large in my thinking about the Yamnaya or its descendants, because it does not occur in any of the tested samples from Yamnaya, Andronovo, Catacomb, Corded War or Bell Beaker, and there is just one M from the Tarim Basin mummies.

It still makes one wonder, as now the earliest identified Y-R1a is mtDNA M.

Related to: "aDNA from Early Bronze Age Cultures of the North Pontic Steppe," M also shows up in the Ukraine.
"14 individuals from three collective burials from the Neolithic Dnieper-Donetz culture" 3 were C (21%)
"three individuals from Bronze Age Kurgan burials" All 3 were C.

Jean M
03-05-2015, 10:36 PM
It still makes one wonder, as now the earliest identified Y-R1a is mtDNA M.

You seem to be counting C, a descendant of M, as M. This is not normal usage. Normal usage is to refer to C as C, D as D etc. All mtDNA haplogroups descend from L. It would be pretty confusing if we just called them all L.

parasar
03-06-2015, 12:23 AM
You seem to be counting C, a descendant of M, as M. This is not normal usage. Normal usage is to refer to C as C, D as D etc. All mtDNA haplogroups descend from L. It would be pretty confusing if we just called them all L.

C is as M as many of the numerous M descendants in South Asia called M. As many lines especially those related to the Americas were studied first they got their own letters. ABCD of which BA are under N.

Jean M
03-06-2015, 12:51 AM
C is as M as many of the numerous M descendants in South Asia called M. As many lines especially those related to the Americas were studied first they got their own letters.

Yes indeed C is a descendant of M. It is also a descendant of L3 and L. But it would be enormously helpful if you would refer to the haplogroup that you mean by the name that is in current use, rather than by the name of an ancestor. My name is Jean. It would confuse people if I suddenly, without warning or explanation, called myself Frederick on the grounds that I am the descendant of one so named.

parasar
03-06-2015, 03:44 AM
Yes indeed C is a descendant of M. It is also a descendant of L3 and L. But it would be enormously helpful if you would refer to the haplogroup that you mean by the name that is in current use, rather than by the name of an ancestor. My name is Jean. I would confuse people if I suddenly, without warning or explanation, called myself Frederick on the grounds that I am the descendant of one so named.

You are missing the point - this is the earliest M in Europe, that is why I called it M while also clearly identifying the variety of M as C. If* wouldn't have called it M had it been found in India.
If you find Y-A in ancient Europe you wouldn't be worrying which variety of A. We often refer to haplogroups in short-form such as R1a (or even R, or P) unless there is need to distinguish what type R1a.

*Edit: I

GailT
03-06-2015, 03:58 AM
this is the earliest M in Europe, that is why I called it M while also clearly identifying the variety of M as C. If wouldn't have called it M had it been found in India.

You also confused me - I assumed M* when you said M, and I thought maybe there was a new M* sample that I missed. It's helpful to name the sample with it's most specific subclade. So I agree with Frederick on this point. :)

parasar
03-06-2015, 04:09 AM
You also confused me - I assumed M* when you said M, and I thought maybe there was a new M* sample that I missed. It's helpful to name the sample with it's most specific subclade. So I agree with Frederick on this point. :)

Okay.
But just to be clear on my query, which mentions C four times, I will repeat it here.

Do you suppose the M lines came in with R1a?
"three Neolithic and all three Bronze Age individuals belonged to mtDNA haplogroup C, which is common in East Eurasian, particularly South Siberian, populations but exceedingly rare in Europe ... Ni58, Ya45, and Ya34 are members of haplogroup C, which is one of the most common mtDNA lineages in East Eurasia and likely originated in South Siberia ... All three Bronze Age samples (D1.8, L8, and L15) also belong to haplogroup C, bringing the total number of haplogroup C individuals in the present study to six ... thought to have originated in South Siberia in the area surrounding Lake Baikal "

Jean M
03-06-2015, 12:11 PM
But just to be clear on my query, which mentions C four times, I will repeat it here.

It is so obvious that mtDNA C arrived in the Urals and European steppe from the Asian steppe that it has already been remarked upon in at least one thesis (which you are aware of) and more than one paper (which you are also aware of), not to mention Ancestral Journeys (2013), p. 64.

So why the query? You want to tie it specifically to one branch of Y-DNA R1? Doubtful, I'd say. MtDNA C and C4a2 were present in Dnieper-Donets farmers, who descended from Dnieper-Donets pottery-making foragers. So it looks like they had taken pottery-making wives from the incomers around Samara.

C5 was even found among Körös farmers in Hungary c. 5500 BC. That is the same date as the C1g at Yuzhnyy Oleni Ostrov. People do get about. :)

Hector
03-06-2015, 02:12 PM
That was nitpicking of the worst kind. Parasar made a quote right bellow that indicated that those M's were C and he was previously using the term "M-line" just a few posts ago.
You guys are simply not happy with the implication that R1a may have an origin farther east.
Anyway you guys are hilarious. The idea of various sub-branches of M526 making independent journeys from as far West as Iran to SE Asia and Oceanea just so that their European descendents are not offended... That is why you are amateurs.

newtoboard
03-07-2015, 12:02 AM
That was nitpicking of the worst kind. Parasar made a quote right bellow that indicated that those M's were C and he was previously using the term "M-line" just a few posts ago.
You guys are simply not happy with the implication that R1a may have an origin farther east.
Anyway you guys are hilarious. The idea of various sub-branches of M526 making independent journeys from as far West as Iran to SE Asia and Oceanea just so that their European descendents are not offended... That is why you are amateurs.

Further east does not include SE Asia.

Jean M
03-07-2015, 12:34 AM
That was nitpicking of the worst kind. Parasar made a quote right below that indicated that those M's were C and he was previously using the term "M-line" just a few posts ago.

You may have immediately understood that Parasar was making a subtle reference to the theory of a South East Asian origin for Y-DNA K2, but I did not. I went to check my online table to see where I might have missed an M. :biggrin1: How shockingly short-sighted and prosaic of me.

Hector
03-07-2015, 01:16 AM
Further east does not include SE Asia.

But definitely in that direction with the ultimate implication that K2b's origin or even K2's origin is in that region. Eastern South Asia is about as far West as any professionals are willing to go whereas amateurs will cling to the hope that it was in West Asia and an entire group of M526+ zombies except K2b2b-M45 made an arduous 2000Km journey to the East with "Go East" stamped on their foreheads by the righteous lightning from the almighty.

newtoboard
03-07-2015, 01:26 AM
But definitely in that direction with the ultimate implication that K2b's origin or even K2's origin is in that region. Eastern South Asia is about as far West as any professionals are willing to go whereas amateurs will cling to the hope that it was in West Asia and an entire group of M526+ zombies except K2b2b-M45 made an arduous 2000Km journey to the East with "Go East" stamped on their foreheads by the righteous lightning from the almighty.

If you look at my past posts discussing this with alan I am one of the few who favored a SE Asian origin for K and P. But this has nothing to do with R1/R1a. Malta is a dead lineage so there was probably a lot of R lineages found in Central Asia and the more westernly ones led to R1 imo.

Jean M
03-07-2015, 01:50 AM
If you look at my past posts discussing this with alan I am one of the few who favored a SE Asian origin for K and P.

You mean over here? http://www.anthrogenica.com/showthread.php?2653-Awesome-new-study-on-the-phylogeny-of-macrohaplogroup-K-Y-DNA/

newtoboard
03-10-2015, 12:51 AM
I wonder if something like this makes sense for post Yamnaya cultures in terms of R1a clades

Poltavka->Andronovo=R1a-Z93(xZ2122)
Poltavka->Abashevo->Timber Grave=R1a-Z2122
Corded Ware(???)->Catacomb->Cimmerian associated cultures(Chernogorovka and Novocherkask)=R1a-Z282 and the rare subclades of R1a-Z280/R1a-M458 found in the North Caucasus

With R1b-Z2103+ being present in Timber Grave and the Cimmerian cultures. And I2a-DIN and I2c2 in the Cimmerian cultures.

Also i wonder if Scythian farmers were of Thraco-Dacian stock or were Slavic speakers.

alan
03-10-2015, 01:14 AM
Following up this earlier discussion: we now have 34 Yamnaya samples yielding mtDNA: (Details with locations: 3961 )

Mesolithic type mtDNA (total 8):
U
U2e1a
U4a1
U5a1a1 (2 samples)
U5
U5a1 (2 samples)

Neolithic type mtDNA (total 26)
H (7 samples)
H2b
H6a1b
H13a1a1a
I
J
K (2 samples)
N1a
T (2 samples)
T1
T1a (2 samples)
T2c1a2
T2a1b1a
W
W3a1a
W6
X

so if they are more than 3 times as much farmer than hunter on the female side that kind of lends further support that the EHG is very strong on the male side.

parasar
03-10-2015, 04:46 PM
You may have immediately understood that Parasar was making a subtle reference to the theory of a South East Asian origin for Y-DNA K2, but I did not. ...

Ultimately perhaps but not proximate. Here I was just thinking about the origin of R1a. As MA-1 is already a pre-R derivative at about the same age as R1a, we do not have to go beyond eastern Asia. Y-R1a with mtDNA M-C is a strong indicator to me about the origins of R1a near NW China, Baikal, S. Siberia.

As I have mentioned before I thought MA1 was admixed for a few reasons:
1. mt-U in MA1.
2. No connection to the Gedrosia type reflection (as seen in MA1) for Amerindians.
3. Amerindians maintaining their relationship to WHG at higher Ks when Papuan and then South Asian split.

It is quite possible that the mtDNA counterpart of early Y-P(Q,R) could have been some kind of mt-M.
http://arslanmb.org/ArmenianDNAProject/mtDNAPhylogeny.jpg


Of the above, the following M possibilities come to mind - C, D - as they are also seen in Amerindians.

As to the migration of mtDNA C to NE Europe,
Ancient DNA Reveals Prehistoric Gene-Flow from Siberia in the Complex Human Population History of North East Europe
http://journals.plos.org/plosgenetics/article?id=10.1371/journal.pgen.1003296
"‘Central/East Siberian’ cluster were predominantly composed of hgs A, B, C, D, F, G, Y, and Z, while in contrast populations of the ‘European’ cluster were characterized by higher frequencies of hgs H, HV, V, U, K, J, T, W, X, and I ...


And in NE Europe we see this C with R1a. So it is not implausible that the Karelian R1a ancestral line too moved from the Baikal/S. Siberia to NE Europe.

ADW_1981
03-10-2015, 05:44 PM
With R1b-Z2103+ being present in Timber Grave and the Cimmerian cultures. And I2a-DIN and I2c2 in the Cimmerian cultures.



I may be completely incorrect, but I'm skeptical that I2 was that far east in those days. I guess it would all depend on the route that YDNA I entered Europe -whether it was Anatolia or northern Eurasia.

Megalophias
03-10-2015, 08:21 PM
I may be completely incorrect, but I'm skeptical that I2 was that far east in those days. I guess it would all depend on the route that YDNA I entered Europe -whether it was Anatolia or northern Eurasia.

I is probably around 30 000 years old, I2 around 4000 years younger. I2a seems to split up right after I2, and younger subclades around the LGM. The Gravettian people of Central Europe went well into Western Russia and could easily have carried it there, unless it was bottled up somewhere until after the LGM. I don't think it really matters whether it came from the east or the south.

R.Rocca
03-13-2015, 10:53 PM
David has run ADMIXTURE to determine Yamnaya related ancestry. Yamnaya kurgan sample I0429, which is R1b-Z2105+ and one of the older Yamnaya samples tested (~3300 BC) is completely lacking in David's Middle Eastern component. Of course we still need ancient DNA samples tested from the Caucasus and the Middle East, but this further puts into doubt what some have suggested...that R1b's movement was just as likely to be from south to the steppe than the other way around. We already knew that R1a and R1b hunter-gatherers were in western Russia for a long time in a non-admixed form and now we have this sample that looks like a non-admixed L23+ sample.

http://eurogenes.blogspot.com/2015/03/yamnaya-related-ancestry-proportions-in.html

David's run also produced the opposite for Early Spanish Neolithic R1b sample I0410, which has zero Yamnaya ancestry. Based on this sample's components, my suggestion that this may be an R1b-V88 sample is a good bet. R1b-V88 likely migrated out of the R1b homeland thousands of years earlier than the Copper Age and had nothing to do with modern L23 lineages, instead spreading both north and south of the Mediterranean.

GailT
03-14-2015, 03:27 PM
That was nitpicking of the worst kind. Parasar made a quote right bellow that indicated that those M's were C and he was previously using the term "M-line" just a few posts ago. You guys are simply not happy with the implication that R1a may have an origin farther east. Anyway you guys are hilarious.

It was not at all clear to me what he meant, and I have absolutely no opinion on where R1a originated. I was commenting on nomeclature for mtDNA, not origins of y-DNA, It seems like you are looking for disagreements where they don't exist. But I'm glad we are all having a good laugh. :)

Jean M
03-14-2015, 04:33 PM
this further puts into doubt what some have suggested...that R1b's movement was just as likely to be from south to the steppe than the other way around.

I gave up the latter idea entirely once we had Haak et al 2015. For the second edn of AJ, I go for an alternative which I mentioned on the forum a while ago, whereby the split on/near the Caspian was with just R1b-V88 going south and the rest of R1b sticking to the north side.


David's run also produced the opposite for Early Spanish Neolithic R1b sample I0410, which has zero Yamnaya ancestry. Based on this sample's components, my suggestion that this may be an R1b-V88 sample is a good bet. R1b-V88 likely migrated out of the R1b homeland thousands of years earlier than the Copper Age and had nothing to do with modern L23 lineages, instead spreading both north and south of the Mediterranean.

Agree again.

Megalophias
03-14-2015, 04:42 PM
David has run ADMIXTURE to determine Yamnaya related ancestry. Yamnaya kurgan sample I0429, which is R1b-Z2105+ and one of the older Yamnaya samples tested (~3300 BC) is completely lacking in David's Middle Eastern component. Of course we still need ancient DNA samples tested from the Caucasus and the Middle East, but this further puts into doubt what some have suggested...that R1b's movement was just as likely to be from south to the steppe than the other way around. We already knew that R1a and R1b hunter-gatherers were in western Russia for a long time in a non-admixed form and now we have this sample that looks like a non-admixed L23+ sample.


Just because Eurogenes new Yamnaya component is 100% in one sample doesn't make it "pure". In fact it almost certainly is recently and heavily admixed, and it clearly isn't an unadmixed descendent of EHG. I don't see that this makes the slightest bit of difference to the question.

There does appear to be a rather strong Central/South Asian element to Yamnaya that may be relevant, however.

Arbogan
03-15-2015, 03:53 AM
Just because Eurogenes new Yamnaya component is 100% in one sample doesn't make it "pure". In fact it almost certainly is recently and heavily admixed, and it clearly isn't an unadmixed descendent of EHG. I don't see that this makes the slightest bit of difference to the question.

There does appear to be a rather strong Central/South Asian element to Yamnaya that may be relevant, however. Despite of what people say the gedrosian/west asian component emerges consistently when created from data sets based on real population data. The data basically seperates different genotypes in humans via matrix calculations into clusters. Correct me if i misunderstood. But say you want to create a k2 progam. You take pathans, tajiks, brahuis on one hand and germans , English and danes. From this data set of 150k snps per population patterns emerge an ancestrally meaningful division of genotypes. north europeans will form one cluster and SC asians in one cluster. Based on their shared ancestral interrelatedness. So saying yamnaya has no SC asian admixture is nonsense. Because that 25 to 21% is consistently based on real patterns that emerge in populations. SNP sequences which form in clusters that are unique to the populations used in datasets. You can argue whether it was Yamnaya which contributed it to ancestors of modern SC asians or vice versa. But based on archaelogical relationships its likely that vice versa also happened. Or that they share a third party contributor. But the fact of the matter is that there is a real relationship.


Whether people like it or not. People are still limiting their approach to the contemporary world conceptions

Hando
03-16-2015, 03:07 AM
I gave up the latter idea entirely once we had Haak et al 2015. For the second edn of AJ, I go for an alternative which I mentioned on the forum a while ago, whereby the split on/near the Caspian was with just R1b-V88 going south and the rest of R1b sticking to the north side.



Agree again.

Are you saying that you think R1b spread from south to the steppe? Ie. From the Caspian to the steppe?

Jean M
03-16-2015, 03:50 AM
Are you saying that you think R1b spread from south to the steppe? Ie. From the Caspian to the steppe?

Oh dear. You picked up my ambiguity too late for me to correct my post for absolute clarity. By 'the latter idea' I meant the idea that R1b had moved from the south of the Caspian via Anatolia to the steppe with dairy farming, which was the guess that went into AJ 1st edition. That idea has been dropped for AJ 2nd edition, which goes for the alternative possibility which I laid out on the forum a while ago, whereby the split on/near the Caspian was with just R1b-V88 going south and the rest of R1b sticking to the north side. R1b-V88 had a different trajectory from the rest of R1b and did not (it seems) enter the steppe at all.

alan
03-16-2015, 07:58 AM
David has run ADMIXTURE to determine Yamnaya related ancestry. Yamnaya kurgan sample I0429, which is R1b-Z2105+ and one of the older Yamnaya samples tested (~3300 BC) is completely lacking in David's Middle Eastern component. Of course we still need ancient DNA samples tested from the Caucasus and the Middle East, but this further puts into doubt what some have suggested...that R1b's movement was just as likely to be from south to the steppe than the other way around. We already knew that R1a and R1b hunter-gatherers were in western Russia for a long time in a non-admixed form and now we have this sample that looks like a non-admixed L23+ sample.

http://eurogenes.blogspot.com/2015/03/yamnaya-related-ancestry-proportions-in.html

David's run also produced the opposite for Early Spanish Neolithic R1b sample I0410, which has zero Yamnaya ancestry. Based on this sample's components, my suggestion that this may be an R1b-V88 sample is a good bet. R1b-V88 likely migrated out of the R1b homeland thousands of years earlier than the Copper Age and had nothing to do with modern L23 lineages, instead spreading both north and south of the Mediterranean.

How did I miss that - a no-middle eastern Z2103 Yamanaya guy. That date of 3300BC is also the normal starting date of true Yamnaya. It does put in mind then the possibility that it was the late Maykop culture that brought the middle eastern component and that this mixing only commenced after 3500BC - hence you could still have early Yamnaya guys with no admixture c. 3300BC.

Generalissimo
03-16-2015, 08:20 AM
All of the Yamnaya are more or less 50% Georgian-like, even that unmixed guy from the K6.

But the mixing happened earlier than 3300 BC and was followed by drift among the Yamnaya population and then a rapid expansion out of the steppe, which is now being picked up as a Yamnaya-specific component. The reason that guy looks unmixed is pobably because he's a prime example of this whole process.

Hando
03-16-2015, 11:16 AM
All of the Yamnaya are more or less 50% Georgian-like, even that unmixed guy from the K6.

But the mixing happened earlier than 3300 BC and was followed by drift among the Yamnaya population and then a rapid expansion out of the steppe, which is now being picked up as a Yamnaya-specific component. The reason that guy looks unmixed is pobably because he's a prime example of this whole process.
By drift, do you mean that the middle eastern type component/admixture was minimized in Yamnaya?

Generalissimo
03-16-2015, 11:30 AM
By drift, do you mean that the middle eastern type component/admixture was minimized in Yamnaya?

http://en.wikipedia.org/wiki/Genetic_drift

Arbogan
03-16-2015, 02:33 PM
All of the Yamnaya are more or less 50% Georgian-like, even that unmixed guy from the K6.

But the mixing happened earlier than 3300 BC and was followed by drift among the Yamnaya population and then a rapid expansion out of the steppe, which is now being picked up as a Yamnaya-specific component. The reason that guy looks unmixed is pobably because he's a prime example of this whole process.
Why then do they appear on oracle as a mixture between tajiks and finns. Their near eastern affinity consistently looks SWC asian.

Generalissimo
03-16-2015, 02:52 PM
Why then do they appear on oracle as a mixture between tajiks and finns. Their near eastern affinity consistently looks SWC asian.

And you think Tajiks and/or Finns got their Near Eastern admixture from SWC Asia, wherever that is? If so, what do you base that on?

Jean M
03-16-2015, 02:56 PM
It does put in mind then the possibility that it was the late Maykop culture that brought the middle eastern component and that this mixing only commenced after 3500BC.

Natalia Shishlina has taken a look at mutual relationships and mutual enrichments of the Steppe and the Caucasus populations throughout the time period from 3800 calBC to 2200 calBC, with a particular focus on population mobility between the two adjacent areas in "The Steppe and the Caucasus During the Bronze Age: Mutual Relationships and Mutual Enrichments", in Counterpoint: Essays in Archaeology and Heritage Studies in Honour of Professor Kristian Kristiansen, BAR International Series 2508 (2013). [There is a copy online, but not easy to link to. Best to Google for the pdf.]


A colonization process: The sedentary Majkop population that lived in a narrow piedmont strip of the North Caucasus subsisted on dry farming and animal husbandry (cattle, sheep and goats) (Trifonov 2002). Small Majkop settlements with an occupation layer of up to 40 cm were located in river valleys.
River pastures and steppe watershed grasslands were used for grazing animals, i.e. horses and pigs, near settlements. Most likely, high mountainous areas were not exploited (Trifonov 2002). Various sub-sectors of sedentary livestock-raising could develop in several landscape areas, such as pig- and sheep-raising, or a
mixed type of pastoralism where cattle, sheep and goats were reared (Munchaev 1975). A dry and warm climate and mild winters with only a small amount of snow supported this type of economy. Though the Majkop population of the North Caucasus is considered to be sedentary, we may suppose that the steppe
environment contributed to their efforts to colonize the still vacant steppe areas. Majkop groups appeared in the Steppe starting from the Early Majkop culture, i.e. 3600-3500 calBC.

Arbogan
03-16-2015, 03:31 PM
And you think Tajiks and/or Finns got their Near Eastern admixture from SWC Asia, wherever that is? If so, what do you base that on?
I dont know maybe because the neolithic sites in zagros and south eastern anatolia are a good deal older than those in the caucasus. And that tajiks and south west central asians in general always end up having a contribution by west eurasian genetic cluster that peaks in in brahuis and other ethnicities in transitional areas along side the borders of the iranian plateau. It also is differentiated from the neolithic signatures found in anatolia and caucasus. Its definately related to neolithic expansions that went eastward but not from the same source responisble for the immigration waves that headed for caucasus. Its nonsense to assume yamnaya received a caucasus based contribution. Not only because of the geographical distance. But there is no archaelogical evidence of a neolithic era caucasus culture being present or having exchanges in the yamnayas found in west kazakhistan . While there is a strong case for BMAC. This why Yamnayas appear 1/4th south west central asian on admixture tests. its also the reason for why you can find those signatures all the way to britain and why north east caucasians that have an elevated affinity to that same ancestry are much less distant than georgians, armenians to yamnaya. Armenians and georgians are probably good examples (especially the latter) of largely unadmixed resorvoirs of pre invasion caucasus.

Megalophias
03-16-2015, 03:53 PM
Its nonsense to assume yamnaya received a caucasus based contribution. Not only because of the geographical distance. But there is no archaelogical evidence of a neolithic era caucasus culture being present or having exchanges in west kazakhistan . While there is a strong case for BMAC.

The Caucasus is directly adjacent to the south Russian steppe. Caucasian interaction with steppe cultures is archaeologically attested in the fourth millennium BC. Why do we need Neolithic Caucasian cultures to be directly interacting with west Kazakhstan?

BMAC is like a thousand years after Yamnaya. Are you talking about Kopet Dag-related cultures? What evidence connects them to the East European steppe during the Neolithic?

Silesian
03-16-2015, 04:09 PM
By drift, do you mean that the middle eastern type component/admixture was minimized in Yamnaya?
You might want to review this video, its quite good. Basically the isotopes in the diet show fish, probably when they were confined to the earliest hunter gatherer period in river valleys; then shift, probably what Generalissimo is picking up in the Georgian/Yamnaya mix. None of the Yamnaya samples were positive for the lactose gene. They were lacking nutrients and living in a violent society. He brings all this out @ 41:32 of this lecture, showing isotopes and speculating they lacked the ability to digest milk, & lactose gene[he was proven right]

https://www.youtube.com/watch?v=d864bwyCAoA

Jean M
03-16-2015, 04:16 PM
. But there is no archaelogical evidence of a neolithic era caucasus culture being present or having exchanges in the yamnayas found in west kazakhistan . While there is a strong case for BMAC.

There is a strong case for Andronovo (ca. 2000–900 BCE) exchanges with the BMAC (ca. 2300–1700 BCE), but the BMAC is too late have any effect on Yamnaya.

DMXX
03-16-2015, 04:25 PM
^ 100% agreed. I posted about this recently (http://www.anthrogenica.com/showthread.php?3680-Yamna-Yamnaya-Y-DNA&p=64134&viewfull=1#post64134). Agricultural terms associated with the BMAC language are found in Asian IE branches only (various Indo-Iranian languages, also in Tocharian A specifically I believe, though IIRC an East Iranic proxy is often invoked). That clearly suggests any BMAC involvement with IE is both in territories east of the Urals and later than Yamnaya.

Arbogan
03-16-2015, 04:36 PM
The Caucasus is directly adjacent to the south Russian steppe. Caucasian interaction with st
eppe cultures is archaeologically attested in the fourth millennium BC. Why do we need Neolithic Caucasian cultures to be directly interacting with west Kazakhstan?
Thats where they excavated the samples. Western branches of yamnaya probably had contact with caucasus. But the samara yamnaya look like a mixture of uralo-baltics with something south west central asian thrown in. Definitely not half caucasus as some proclaim. Theyre not good examples of western PIE which is why they genetically resemble mordvins rather than russians or ukranians. The farmer ancestry they have looks gedrosian

The Caucasus is BMAC is like a thousand years after Yamnaya. Are you talking about Kopet Dag-related cultures? What evidence connects them to the East European steppe during the Neolithic?
The samara yamnaya were excavated in an area just north of kazakhistan. Not the pontic steppe or ukraine. And based on their signatures they seem to have a swc asian affinity. Now if its the kopet dag culture or some unknown older culture im not sure of. But BMAC is defintely a later continuation of that cultural influence. And I dont see why not. There were post neolithic sprawling all over the western area of the iranian plateau at the same time as these yamnaya were dated . Some of them have clear SWC asian affinities.

Megalophias
03-16-2015, 04:48 PM
Thats where they excavated the samples. Western branches of yamnaya probably had contact with caucasus. But the samara yamnaya look like a mixture of uralo-baltic with something south west central asian thrown in. Definitely not half caucasus as some proclaim.
The samara yamnaya were excavated in an area just north of kazakhistan. Not the pontic steppe or ukraine. And based on their signatures they seem to have swc asian affinity.

Yes, they were from near Kazakhstan. It is obviously geographically plausible that they were related to Central Asians. They were also part of the Yamnaya horizon, which had contact with the Caucasus, so it's also obviously possible that they could have been connected to the Caucasus. I was asking you for the archaeological evidence that connects SWC Asia to the East European steppe at this period.

Your opinion is that they can't be half Caucasian, but profesional geneticists have fancy models that say they can be half Caucasian. Do you have some kind of actual evidence beyond eyeballing admixture components? Do you happen to magically know what sort of ancestry Chalcolithic Caucasians and Chalcolithic Central Asians had?

This is not to say there isn't some kind of South Asian-like affinity in the Yamnaya guys, there sure seems to be, but we don't know what the distribution of that kind of ancestry was like back then. And of course South Asian-like ancestry is widespread in West Asia at present.

Arbogan
03-16-2015, 05:08 PM
Yes, they were from near Kazakhstan. It is obviously geographically plausible that they were related to Central Asians. They were also part of the Yamnaya horizon, which had contact with the Caucasus, so it's also obviously possible that they could have been connected to the Caucasus. I was asking you for the archaeological evidence that connects SWC Asia to the East European steppe at this period.

Your opinion is that they can't be half Caucasian, but profesional geneticists have fancy models that say they can be half Caucasian. Do you have some kind of actual evidence beyond eyeballing admixture components? Do you happen to magically know what sort of ancestry Chalcolithic Caucasians and Chalcolithic Central Asians had?

This is not to say there isn't some kind of South Asian-like affinity in the Yamnaya guys, there sure seems to be, but we don't know what the distribution of that kind of ancestry was like back then. And of course South Asian-like ancestry is widespread in West Asia at present.

Instead of being a snappy anglo ass(my mistake, sorry greek members. This why you shouldn't have a ethnic pseudonym). You could cite the fancy methadology of haak et al. Because f3 statistics , tree mix and other tools have their limitations. I once got a run where i showed up as having 50% bedouin-like contribution which is impossible. I dont get why admixture gets a flak. Its based on genotypic patterns.

Silesian
03-16-2015, 05:23 PM
Yes, they were from near Kazakhstan. It is obviously geographically plausible that they were related to Central Asians. They were also part of the Yamnaya horizon, which had contact with the Caucasus, so it's also obviously possible that they could have been connected to the Caucasus. I was asking you for the archaeological evidence that connects SWC Asia to the East European steppe at this period........
Eurogenes . K15 show Yamnaya I0429 Sok River, Samara, Russia 3339-2917 calBCE [I][lat-53.38] [long-50.39]
[North_Sea-29.55_25.51][Atlantic-1.62][Baltic-12.2_17.68][Eastern_Euro-26.14_41.73][West_Med-0][West_Asian-17.96][East_Med-0][Red_Sea-0][South_Asian-7.37_1.99][Southeast_Asian-0][Siberian-0][Amerindian-5.16_12.02][Oceanian-0_1.07][Northeast_African-0][Sub-Saharan-0]
Keep in mind that R1b IO429 is only about 2+/- miles from the 7650+/- sample was taken from.[Samara_HG I0124 Sok River, Samara, Russia 5650-5555 calBCE [lat-53.4] [lon-50.4] Samara_HG I0124 Sok River[/COLOR]1d R1b1aU5a
https://docs.google.com/spreadsheets/d/1QPTmyarOBBEZfXnLI5L64ueJNG34jgy4QgQ_1nSYtnM/edit?pli=1
https://docs.google.com/spreadsheets/d/1fKE6hbeXUWYEUZIBiU8QKIktHmNNC5NE4wJk3NsTuyY/edit?pli=1#gid=1448840466

Megalophias
03-16-2015, 05:41 PM
Instead of being a snappy anglo ass(my mistake, sorry greek members. This why you shouldn't have a pseudo ethnic ethonym). You could cite the fancy methadology of haak et al. Because f3 statistics , tree mix and other tools have their limitations. I once got a run where i showed up as having 50% bedouin-like contribution which is impossible. I dont get why admixture gets a flak. Its based on genotypic patterns.

Sorry for being snappy. I am genuinely interested here, I actually want to know what the archaeological connection between S-C Asia and the East European steppe during this period is.

I'm not saying that Haak et al are necessarily right and whatever you are basing your opinion on is necessarily wrong. I'm asking why we should believe you, random guy on the Internet, when you say they are "definitely not half Caucasian", over the latest findings in the scientific literature. Could you link to whatever your evidence actually is?

(I'm fairly sure that "Gigantic snake" is not a real Greek name. It is actually the scientific name of a non-existent genus.)

Arbogan
03-16-2015, 06:01 PM
Eurogenes . K15 show Yamnaya I0429 Sok River, Samara, Russia 3339-2917 calBCE [I][lat-53.38] [long-50.39]
[North_Sea-29.55_25.51][Atlantic-1.62][Baltic-12.2_17.68][Eastern_Euro-26.14_41.73][West_Med-0][West_Asian-17.96][East_Med-0][Red_Sea-0][South_Asian-7.37_1.99][Southeast_Asian-0][Siberian-0][Amerindian-5.16_12.02][Oceanian-0_1.07][Northeast_African-0][Sub-Saharan-0]
Keep in mind that R1b IO429 is only about 2+/- miles from the 7650+/- sample was taken from.[Samara_HG I0124 Sok River, Samara, Russia 5650-5555 calBCE [lat-53.4] [lon-50.4] Samara_HG I0124 Sok River1d R1b1aU5a
https://docs.google.com/spreadsheets/d/1QPTmyarOBBEZfXnLI5L64ueJNG34jgy4QgQ_1nSYtnM/edit?pli=1
https://docs.google.com/spreadsheets/d/1fKE6hbeXUWYEUZIBiU8QKIktHmNNC5NE4wJk3NsTuyY/edit?pli=1#gid=1448840466

Interesting. I wonder if we'll see runs with Project members. k15 and k6 also corroborates Yamnaya being PIE and proto uralic.

Generalissimo
03-16-2015, 08:18 PM
Natalia Shishlina has taken a look at mutual relationships and mutual enrichments of the Steppe and the Caucasus populations throughout the time period from 3800 calBC to 2200 calBC, with a particular focus on population mobility between the two adjacent areas in "The Steppe and the Caucasus During the Bronze Age: Mutual Relationships and Mutual Enrichments", in Counterpoint: Essays in Archaeology and Heritage Studies in Honour of Professor Kristian Kristiansen, BAR International Series 2508 (2013). [There is a copy online, but not easy to link to. Best to Google for the pdf.]

Interesting quote from that paper in the context of the uniparental markers found in the Yamnaya.


It appears that two young women of 14-17 and 20 years old were born somewhere near the coastlines of the Caspian or Black Seas (North Caucasus region) and later married and moved to the steppe.

Page 58

Jean M
03-16-2015, 09:15 PM
Interesting quote from that paper in the context of the uniparental markers found in the Yamnaya.
It appears that two young women of 14-17 and 20 years old were born somewhere near the coastlines of the Caspian or Black Seas (North Caucasus region) and later married and moved to the steppe.

:) What was the translator thinking? Possibly there is not a lot of gay marriage where he/she comes from. So the word 'men' after 'marriage' was deemed unnecessary.

But seriously. Yes the work done with isotopes was what struck me. Clever stuff. I just wish excavators could always pay for aDNA as well.

alan
03-16-2015, 11:25 PM
Interesting paper about women moving from sea short to steppe. It is interesting that they were eating a diet with a significant marine aspect as in general early farmers tended in many parts of Europe to not eat seafood. Would be curious to know which cultures of these periods around the Pontic-Caspian had a diet with a big marine component. I do recall hearing about heavily fishing based cultures on Azov in the Neolithic.

newtoboard
03-17-2015, 12:50 AM
Interesting paper about women moving from sea short to steppe. It is interesting that they were eating a diet with a significant marine aspect as in general early farmers tended in many parts of Europe to not eat seafood. Would be curious to know which cultures of these periods around the Pontic-Caspian had a diet with a big marine component. I do recall hearing about heavily fishing based cultures on Azov in the Neolithic.


Either way this would seem to support the idea that DMXX and I have that Samara Yamnaya is a poor fit for Asian IE speakers and the population that brought Indo-Iranian languages to Asia will be more southern and have more farmer ancestry (and in my opinion more ANE as well). If I had to guess something like 50% ANE, 40% ENF, and 10% WHG.

The Azov and Don/Donets region is so hard to figure out. I have no idea what was there before today's Slavic speakers the Crimean Khanate and Timber Grave/Scythians. I don't really have a a strong opinion/ idea on whether R1a or R1b dominated this region during Catacomb.

It is clear that Catacomb features start to form and deveop in situ during Yamnaya and then take off in Catacomb. During Catacomb , Maykop contacts are intensified. And Catacomb shows links with Corded Ware. So was there a migration from Corded Ware and/or Maykop to Catacomb? Does this cause the apparent discontinutiy? Is it caused by mtdna or ydna? On the other hand was the migration from Catacomb to Corded Ware (ie Abashevo which appears to end up showing steppe features unlike Fataynovo/Balanovo due to likely Catacomb/Poltavka contacts)? Was R1a/R1b still dominant or will Catacomb end up looking like an Indo-Europeanized North Caucasian culture with men mostly being G2 (remember we have 6 y G2 and mt I from this region later on in the Don region). So many questions and almost no clarity.

R.Rocca
03-17-2015, 01:13 AM
Either way this would seem to support the idea that DMXX and I have that Samara Yamnaya is a poor fit for Asian IE speakers and the population that brought Indo-Iranian languages to Asia will be more southern and have more farmer ancestry (and in my opinion more ANE as well). If I had to guess something like 50% ANE, 40% ENF, and 10% WHG.

The Azov and Don/Donets region is so hard to figure out. I have no idea what was there before today's Slavic speakers the Crimean Khanate and Timber Grave/Scythians. I don't really have a a strong opinion/ idea on whether R1a or R1b dominated this region during Catacomb.

It is clear that Catacomb features start to form and deveop in situ during Yamnaya and then take off in Catacomb. During Catacomb , Maykop contacts are intensified. And Catacomb shows links with Corded Ware. So was there a migration from Corded Ware and/or Maykop to Catacomb? Does this cause the apparent discontinutiy? Is it caused by mtdna or ydna? On the other hand was the migration from Catacomb to Corded Ware (ie Abashevo which appears to end up showing steppe features unlike Fataynovo/Balanovo due to likely Catacomb/Poltavka contacts)? Was R1a/R1b still dominant or will Catacomb end up looking like an Indo-Europeanized North Caucasian culture with men mostly being G2 (remember we have 6 y G2 and mt I from this region later on in the Don region). So many questions and almost no clarity.

Is this from Ancient DNA?

newtoboard
03-17-2015, 10:50 PM
Is this from Ancient DNA?

From some Alanic/Scythian Kurgans (around 600 BC) so I was wondering if this is what the catacomb people were like. A predominantly farmer poulation that was Indo-Europeanized. There was a thread on these samples somewhere in this forum.

newtoboard
03-17-2015, 11:03 PM
Is this from Ancient DNA?

http://www.anthrogenica.com/showthread.php?3251-New-study-Alans-G2*

Arbogan
03-18-2015, 06:50 PM
I've been looking at different yamnaya scores. There seems to be this weird connection between yamnaya, north-Caucasus and central-asia. I noted before that ossetians and north-Caucasians always scored south-central Asian affinity. I assumed that was a consequence of indo-Iranians, and the invasions of other steppe riders(namely turkics) . But now i'm beginning to think of it in another way. Whenever you run yamnaya with a south-Caucasus based K, the results always show a diminished affinity while almost singular affinity to south central asia. However whenever you include a North-Caucasus based K, this affinity splits into north-causcasus and south-central Asian like component. I'm beginning to think whatever this non-EHG west-Eurasian group(that people call Armenian/georgian farmer) that contributed to yamnaya was also related to south-central Asians and north-caucasians. Hence this very close affinity, and why Caucasus and gedrosia always merge as composites in Ks whenever they're not diffrentated. Maykop could have been a representative of such a post-neolithic culture. Whatever people want to call it ENF+ANE or something else.

Dr_McNinja
03-18-2015, 07:32 PM
Either way this would seem to support the idea that DMXX and I have that Samara Yamnaya is a poor fit for Asian IE speakers and the population that brought Indo-Iranian languages to Asia will be more southern and have more farmer ancestry (and in my opinion more ANE as well). If I had to guess something like 50% ANE, 40% ENF, and 10% WHG.

The Azov and Don/Donets region is so hard to figure out. I have no idea what was there before today's Slavic speakers the Crimean Khanate and Timber Grave/Scythians. I don't really have a a strong opinion/ idea on whether R1a or R1b dominated this region during Catacomb.

It is clear that Catacomb features start to form and deveop in situ during Yamnaya and then take off in Catacomb. During Catacomb , Maykop contacts are intensified. And Catacomb shows links with Corded Ware. So was there a migration from Corded Ware and/or Maykop to Catacomb? Does this cause the apparent discontinutiy? Is it caused by mtdna or ydna? On the other hand was the migration from Catacomb to Corded Ware (ie Abashevo which appears to end up showing steppe features unlike Fataynovo/Balanovo due to likely Catacomb/Poltavka contacts)? Was R1a/R1b still dominant or will Catacomb end up looking like an Indo-Europeanized North Caucasian culture with men mostly being G2 (remember we have 6 y G2 and mt I from this region later on in the Don region). So many questions and almost no clarity.
I think it'd be closer to the South Central Asian population that keeps coming up in the calculators as an ancestor of Yamnaya.

Yamnaya: 34 ANE, 28 ENF, 35 WHG

Mystery SC-Asian pop: 31 ANE, 48 ENF, 19 WHG (mixing this with Karelia/Samara HG gives Yamnaya).

So it's related to Yamnaya (since I'm assuming they're jointly linked to a PIE homeland), and seems more like the kind of Indo-Iranian population in South Asia.

However, Yamnaya itself winds up being a better fit for the most WHG carrying North Indians while the more southernly one becomes heavily predominant when fitting to any population west of the Indus (like in Afghanistan). I say this because when using Yamnaya, or another European component, in addition to Gedrosian which itself should eat up all the excess ENF in SC-Asia, it leaves a lot of ENF/"Caucasian" (basically Near East Early Neolithic Farmers) admixture left over in Afghanistan, near the Indus, and in some upper castes of North India (South India as well, but in the South this extra is present across the board and is probably a relic of ancient farmers).

This excess ENF can either be treated as just a basal layer that forms a continuum to the actual Caucasus (and some no doubt is), or as having come with the WHG (and thus from the same Steppe-like population which brought WHG).

Not to mention populations mix slowly and repeatedly and then their descendants mix again, so there were likely multiple waves of increasing ENF decreasing WHG Steppe-like Indo-Iranian populations.

Vadim Verenich
03-22-2015, 07:28 PM
Now look at what what i've done with both datasets

https://plot.ly/~vadimverenich/15/two-first-principal-components/

R.Rocca
04-06-2015, 07:58 PM
David (aka Generalissimo) has posted this in the comments of his Eurogenes blog:


The teal component definitely looks like influence from a Mesolithic population from somewhere, at least in part. I was shown an e-mail recently from David Anthony in which he speculated that this component might be from a Mesolithic population from the southern steppe and/or north Caucasus. He gave his permission to share this info in public, underlining that it was just his opinion.

Btw, this is also where I learned that, unlike the Yamnaya, Khvalynsk samples were only 25% present-day Armenian-like.

Samara/Karelia were both R1+ and had little to no Near Eastern genetic components and Yamnaya were also all R1+ and 50% EHG and 50% Armenian like. So, if these new Khvalynsk Culture samples also wind up being heavily R1+ and being only 25% Armenian like, and being that Khvalynsk Culture falls chronologically in between Samara and Yamnaya, we should get a pretty good picture of how R1 lineages became more Armenian-like through time (and not the other way around). Hopefully these samples will also give us an idea of what Maykop will look like.

Krefter
04-06-2015, 08:50 PM
Samara/Karelia were both R1+ and had little to no Near Eastern genetic components and Yamnaya were also all R1+ and 50% EHG and 50% Armenian like. So, if these new Khvalynsk Culture samples also wind up being heavily R1+ and being only 25% Armenian like, and being that Khvalynsk Culture falls chronologically in between Samara and Yamnaya, we should get a pretty good picture of how R1 lineages became more Armenian-like through time (and not the other way around). Hopefully these samples will also give us an idea of what Maykop will look like.

I don't see how the Khvalynsk guys won't come out R1. From the Mesolithic-Iron age all North Eurasian Y DNA, except for a few(admixture with foreigners?), are R1a or R1b. If Khvalynsk comes out R1b1a-P297 or R1b1a2-M269, there's little debating whether R1b-P297 came from Central Asia-East Europe.

R.Rocca
04-07-2015, 11:48 AM
I don't see how the Khvalynsk guys won't come out R1. From the Mesolithic-Iron age all North Eurasian Y DNA, except for a few(admixture with foreigners?), are R1a or R1b. If Khvalynsk comes out R1b1a-P297 or R1b1a2-M269, there's little debating whether R1b-P297 came from Central Asia-East Europe.

You'd be surprised at what some people will debate :D

rms2
04-07-2015, 12:02 PM
You'd be surprised at what some people will debate :D

Oh, yeah, especially when it comes to R1b and its alleged connection to the Indo-Europeans. I don't think the debate would stop even if every set of ancient male remains throughout Eurasia were exhumed and y-dna successfully extracted from them. It's amazing.

Augustus
04-07-2015, 09:22 PM
Well start guessing as to what clade it is. But I don't see how just any R1 found there concludes the IE debate definitively. It has to depend on what clade.

The gap between the Samara HG and the Yamna samples is far too short for the Samara HG to be considered a direct ancestor to the Yamna folk, no? Doesn't this mean we should expect another R1b1* or some other similar clade to Samara HG?