PDA

View Full Version : Allentoft's Yamna I2a



George
06-10-2015, 10:38 PM
Figure 1 indicates that some of the Yamna individuals (at least 1, but I don't have access to the article) buried in the heartland between Don and Volga were (was) I2a. Any comments?

George
06-12-2015, 03:01 AM
Subject to correction by experts.== It seems to me that the Yamna I2a does not represent "Trypilian" into-steppe migration (which the Allentoft paper apparently accepts). And judging by the spare information given in Figure 2, it is quite different from the Remedello BA I2a, and probably from the Hungarian BA samples. This Yamna I2a does not have any measurable genetic "farmerisms". I would hence think that it is also not of Usatovo descent. But if this is so, then we possibly might have a very interesting situation here. That of a HG who joined the PIE development directly from the HG stage.== We absolutely need further ancient DNA from the areas neglected by the Haak and Allentoft enterprises.

Megalophias
06-12-2015, 05:04 AM
Why shouldn't I2 be native to the region? It had contact with Central Europe in the Paleolithic (Gravettian particularly). Many people have suggested that R came in from the east quite late, in the Late Upper Paleolithic or Mesolithic.

vettor
06-12-2015, 06:27 AM
Why shouldn't I2 be native to the region? It had contact with Central Europe in the Paleolithic (Gravettian particularly). Many people have suggested that R came in from the east quite late, in the Late Upper Paleolithic or Mesolithic.

Which is the older I2 , steppe, hungarian or remendello ones?

George
06-12-2015, 12:45 PM
Why shouldn't I2 be native to the region? It had contact with Central Europe in the Paleolithic (Gravettian particularly). Many people have suggested that R came in from the east quite late, in the Late Upper Paleolithic or Mesolithic.

Anthrogenica experts are very busy decoding the info from Allentoft with respect to those Y markers which are of particular interest to them. Eventually someone will get to the Yamna results. This will give some preliminary answers. I believe Eurogenes' Davidski is also planning some analysis for next week.

eastara
06-14-2015, 12:36 AM
The I2 branches have already been identified:
RISE247 Hungary Vatya I2a2a-L368
RISE254 Hungary Vatya I2a2a-L59

Hungary belong to I2a2a, old I2b. It is mentoined in the paper that during the Middle Bronze age a back migration from West to East is observed. This very well explains the I2b finds among Tartars and as far East as Pakistan.

Regarding the I2 from Italy they are the "Sardianian" I2a1 type found before in Neolithic remains.

RISE486 Italy Remedello I2a1a1a-L672/S327

George
06-14-2015, 05:07 AM
The I2 branches have already been identified:
RISE247 Hungary Vatya I2a2a-L368
RISE254 Hungary Vatya I2a2a-L59

Hungary belong to I2a2a, old I2b. It is mentoined in the paper that during the Middle Bronze age a back migration from West to East is observed. This very well explains the I2b finds among Tartars and as far East as Pakistan.

Regarding the I2 from Italy they are the "Sardianian" I2a1 type found before in Neolithic remains.

RISE486 Italy Remedello I2a1a1a-L672/S327

And it seems (see post#360 in the main Allentoft thread) that the Yamna I2a is L-699. Part of M223 like the two Vatyas, but with no EEF autosomal component.

Christina
06-14-2015, 05:14 AM
Just a thought: it may solve the oft-repeated question as to why uniparental markers in Sardinia (that is, I2a1) "don't match the autosomal DNA" (being EEF), to just imagine that I2a1-M26 was NOT necessarily in Europe during the WHG times in great numbers, but that the mass migrations were of the Cardial people, who had already picked up farming. In other words, that I2a1 is one of the few I2 clades that is not representative of WHG.

Chad Rohlfsen
06-14-2015, 05:40 AM
The I2a is likely Catacomb, not Yamnaya. Ulan IV is a Catacomb burial. Stalingrad Quarry R1b is likely a Catacomb too. I am digging into it.

lgmayka
06-14-2015, 02:38 PM
RISE247 Hungary Vatya I2a2a-L368
YFull lists L368 at the I-M436 level (http://yfull.com/tree/I-M436/), which includes both I-M223 and I-L38. But I don't know whether they specifically found L368+ in a verified I-L38 sample.

RISE486 Italy Remedello I2a1a1a-L672/S327
YFull lists L672 in about 30 different positions on the haplotree, so it's quite unreliable unless M26+ is also present.

In general, these low-quality ancient samples that have no Y-STRs require far greater caution in classification than most of us use in dealing with modern FTDNA accounts. Every SNP should be checked against YFull (or a similar full-Y database) to ensure that it is found in no more than one or two positions on the haplotree--otherwise it is unreliable except perhaps if the context is restricted by other reliable SNPs.

George
06-14-2015, 02:50 PM
YFull lists L368 at the I-M436 level (http://yfull.com/tree/I-M436/), which includes both I-M223 and I-L38. But I don't know whether they specifically found L368+ in a verified I-L38 sample.

YFull lists L672 in about 30 different positions on the haplotree, so it's quite unreliable unless M26+ is also present.

In general, these low-quality ancient samples that have no Y-STRs require far greater caution in classification than most of us use in dealing with modern FTDNA accounts. Every SNP should be checked against YFull (or a similar full-Y database) to ensure that it is found in no more than one or two positions on the haplotree--otherwise it is unreliable except perhaps if the context is restricted by other reliable SNPs.

But in your opinion, the L699 reading for the Yamna I2a does connect it to the larger M223 family?

Jean M
06-14-2015, 03:33 PM
Figure 1 indicates that some of the Yamna individuals (at least 1, but I don't have access to the article) buried in the heartland between Don and Volga were (was) I2a. Any comments?

Hi George. The sample in question is RISE552 from kurgan 4, Ulan IV in Russia. Date: 2849-2143 BC.

Here's the relevant extract from Ancestral Journeys revised edition, in the chapter on Indo-Europeans and genetics:


Genetic fellow-travellers

We should not assume that the Indo-Europeans were all descended from the R1 founder... Certain other haplogroups appear to travel with subclades of R1 in the migrations of Indo-European speakers. A group of Bronze Age skeletons found in Lichtenstein Cave, in Lower Saxony, provide a concrete example of Y-DNA haplogroups mixed within one band. The men included two possibly of Y-DNA R1a1, one of R1b, but no fewer than twelve of I2a2b (L38/S154). The last two haplogroups still reflect the connection shown in the cave. The present-day distributions of I2a2b and R1b-L21 both flow along the Rhine and into the British Isles.

As we have seen, I2a appears in a hunter turned farmer in Hungary c. 5780-5650 BC (p. 64). So Cucuteni-Tripolye farmers may have contributed I2a into the Yamnaya-with-Cucuteni-Tripolye cultural amalgam. If I2a2 was associated with Usatovo and the villages along the Dniester, that would explain why I2a2b (L38/S154) appears alongside R1a1a after apparently migrating up the river and around the Carpathians into present-day Germany (Lichtenstein Cave).

Haplogroup I2a1 (P37.2/PF4004) also appears in early farmers in Hungary. So some I2a1b (M423) carriers could have been living in the Late Cucuteni-Tripolye towns of the Middle Dnieper. If their descendants chose to remain in what became the Proto-Slavic heartland, together with R1a1a men until population growth pushed them outward in all directions in the early Middle Ages, that would explain the pattern we see today. Haplogroups R1a-M458 and I2a1b2a1 (CTS5966) are strongly correlated with the distribution of Slavic languages.

While I agree with:


Why shouldn't I2 be native to the region?

we would then have the problem of explaining why I2 by and large did not spread eastwards into Asia with R1a in the Yamnaya-derived cultures. That is why I favoured a CT farming location. Could be just chance, I suppose.

[Added]

This very well explains the I2b finds among Tartars and as far East as Pakistan.

There is some I2 where we might expect it in Asia.

Jean M
06-14-2015, 03:53 PM
The I2a is likely Catacomb, not Yamnaya. Ulan IV is a Catacomb burial. Stalingrad Quarry R1b is likely a Catacomb too. I am digging into it.

The sample is a late one for Yamnaya, which I do think makes a difference. It gives more time for Yamnaya mixture.

lgmayka
06-15-2015, 12:46 AM
But in your opinion, the L699 reading for the Yamna I2a does connect it to the larger M223 family?
YFull has found L699 in only one place on the haplotree, and gives it a level: I-L699 (http://yfull.com/tree/I-L699/), which is a subclade several levels down within I-M223.

eastara
06-15-2015, 07:24 AM
There were some myths about a lot of haplogroup I in Asia like the 20% among Kurd and Iranians. In fact the only 2 branches that have some sizable membership are I2a2a(I-M223, former I2b1) and I2c.
Here are the only I's found in the more recent and reliable study "Afghan Hindu Kush: Where Eurasian Sub-Continent Gene Flows Converge"
I-M223 AZ6_61 Hazara-Bamiyan
I-M223 IR2_15 Esfahan
I-M223 IR4_16 Khorasan
I-M223 IR7_1 South Iran
I-M223 CEPH_5_A11_HGDP00127 Hazara
I-M258 AZ6_5 Hazara-Bamiyan
I-M423 TJ7_14 Tajik-Takhar
I-M436 MG3_53 Mongol-NorthWest
I-M436 IR7_7 South Iran
I-M438 TJ1_27 Tajik-Balkh
I-M438 TJ1_48 Tajik-Balkh

Unfortunately all I2b1 project administrators and other specialists were preoccupied with the European spread of I2b1 and ignored the occurrence as far East as Pakistan, at most attributing to recent European admixture.
In fact the population group with highest percent (18%) I2b1 are the so called Azov Greeks. In fact "Greek" is only their belonging to the Greek Orthodox church, they speak Turkic language and used to live in the Crimean peninsular. I2b1 is found all along the steppes among Tartars, North Caucasus, Armenians, Iran, Afghanistan, Pakistan, there is even I2a2 M436 in Mongolia! It was not clear before how it got there, but the Bronze era spread of Europeans East leaves no doubt about it.

Haplogroup I2c I-L596 was surprisingly found in Mesolithic remains from Sweden and challenged the Anatolian origin most people supported before. I think it spread the other way around - from Europe and the Balkans to Anatolia. Probably before it was more common there, but the recent exodus of the Greeks from Western Anatolia and the Armenians in Eastern localised it only around South Caucasus.

sparkey
06-15-2015, 11:09 PM
Haplogroup I2c I-L596 was surprisingly found in Mesolithic remains from Sweden and challenged the Anatolian origin most people supported before. I think it spread the other way around - from Europe and the Balkans to Anatolia. Probably before it was more common there, but the recent exodus of the Greeks from Western Anatolia and the Armenians in Eastern localised it only around South Caucasus.

People thought that I2c spread from Anatolia? Even though I2c's more diverse branch (I2c1) is pretty much Europe-exclusive and only the less diverse branch (I2c2) appears significantly outside Europe, and its closest relatives (I2b and then the rest of I2) are also very European? I wonder what the justification was?

sparkey
06-15-2015, 11:17 PM
For those familiar with Nordtvedt's nicknames for I2 subclades, I2-M223>L699 is what he called "Cont3b." It's a subclade of "Cont3," or I2-M223>L701, which was already known from modern samples to have a more eastern inclination than most other I2-M223 subclades. Alongside the typical European distribution in Britain, Netherlands, Germany, etc., modern I2-M223>L699 includes a known presence in Georgia, and I've even seen an Indian sample.

DMXX
06-15-2015, 11:43 PM
People thought that I2c spread from Anatolia? Even though I2c's more diverse branch (I2c1) is pretty much Europe-exclusive and only the less diverse branch (I2c2) appears significantly outside Europe, and its closest relatives (I2b and then the rest of I2) are also very European? I wonder what the justification was?

One of the main justifications was the apparent presence of Y-DNA I in Anatolia via Nasidze et al. (http://onlinelibrary.wiley.com/doi/10.1046/j.1529-8817.2005.00174.x/full), which found 33% and 16% of Zazaki and Kurmanji Kurds respectively to have it. Some later studies picked up the occasional Y-DNA I2 lineage in Anatolia and deeper testing of West Asians (particularly Armenians (https://www.familytreedna.com/groups/armeniadnaproject/about/results)) revealed Y-DNA I2 is present among them in non-trivial frequencies.

For the record, Nasidze et al.'s findings vis-a-vis Kurdish Y-DNA I have not been replicated yet. I should note quite a lot of the Nasidze papers focusing on West Asia have produced Y-DNA frequency results which, in retrospect, appear anomalous when compared to later and more expansive testing in the region. It's anyone's guess whether they were genuine hyper-drifted Y-DNA I, or the result of some sort of lab error.

eastara
06-16-2015, 07:33 AM
People thought that I2c spread from Anatolia? Even though I2c's more diverse branch (I2c1) is pretty much Europe-exclusive and only the less diverse branch (I2c2) appears significantly outside Europe, and its closest relatives (I2b and then the rest of I2) are also very European? I wonder what the justification was?

Sorry, I meant I2c2 PF3827, or the old Cluster B. Exactly that was found in the old European bones. As it was more common among Caucasians, Anatolians and Jews, the odd European members, mainly from South East Europe were considered by many a back migration from there

George
06-16-2015, 01:46 PM
For those familiar with Nordtvedt's nicknames for I2 subclades, I2-M223>L699 is what he called "Cont3b." It's a subclade of "Cont3," or I2-M223>L701, which was already known from modern samples to have a more eastern inclination than most other I2-M223 subclades. Alongside the typical European distribution in Britain, Netherlands, Germany, etc., modern I2-M223>L699 includes a known presence in Georgia, and I've even seen an Indian sample.

It seems (a blog entry in Eurogenes) that the Allentoft Yamna I2a is actually a further split from L-699, to wit: S-12195. Does the above also apply to this more precise identity? The Allentoft autosomals certainly suggest that Yamna I2a 's ancestors were around the area east and north of "farmer types" for a long time if not altogether. Is it impossible at this time to decide whether the West European S-12195 was initially born in Europe or was a later backflow thereto?

Jean M
06-16-2015, 02:00 PM
It seems (a blog entry in Eurogenes) that the Allentoft Yamna I2a is actually a further split from L-699, to wit: S-12195

Where did you see this George? I can't see that. All I have is the report by Smal that RISE552 is I2a-L699 [I2a2a1b1b] at least.
http://www.anthrogenica.com/showthread.php?4640-Population-genomics-of-Bronze-Age-Eurasia-%28Allentoft-et-al-2015%29&p=89228&viewfull=1#post89228

To clarify: I2a2a (M223) > I2a2a1 (CTS616, CTS9183) > I2a2a1b (CTS10057, CTS10100) > I2a2a1b1 (L701, L702) > I2a2a1b1b (L699)

George
06-16-2015, 02:47 PM
Where did you see this George? I can't see that. All I have is the report by Smal that RISE552 is I2a-L699 [I2a2a1b1b] at least.
http://www.anthrogenica.com/showthread.php?4640-Population-genomics-of-Bronze-Age-Eurasia-%28Allentoft-et-al-2015%29&p=89228&viewfull=1#post89228

To clarify: I2a2a (M223) > I2a2a1 (CTS616, CTS9183) > I2a2a1b (CTS10057, CTS10100) > I2a2a1b1 (L701, L702) > I2a2a1b1b (L699)

Hello Jean, It was a short entry by Krefter. He got it from here: https://genetiker.wordpress.com/2015/06/11/y-haplogroups-for-prehistoric-eurasian-genomes/

MT1976
07-07-2015, 01:25 PM
Kristina said:

"Just a thought: it may solve the oft-repeated question as to why uniparental markers in Sardinia (that is, I2a1) "don't match the autosomal DNA" (being EEF), to just imagine that I2a1-M26 was NOT necessarily in Europe during the WHG times in great numbers, but that the mass migrations were of the Cardial people, who had already picked up farming. In other words, that I2a1 is one of the few I2 clades that is not representative of WHG."


Yes it was. I2a1a ('pre-M26) already found in Mesolithic sweden , and Neolithic Paris basin, neither of which have anything to do with Cardial. ..

Jean M
07-07-2015, 04:08 PM
Hello Jean, It was a short entry by Krefter. He got it from here: https://genetiker.wordpress.com/2015/06/11/y-haplogroups-for-prehistoric-eurasian-genomes/

Thanks. I must have missed that entry by Genetiker. I would go with Smal for the moment, as Genetiker is not publicising the details of his reading.

Motzart
07-07-2015, 04:49 PM
One of the main justifications was the apparent presence of Y-DNA I in Anatolia via Nasidze et al. (http://onlinelibrary.wiley.com/doi/10.1046/j.1529-8817.2005.00174.x/full), which found 33% and 16% of Zazaki and Kurmanji Kurds respectively to have it. Some later studies picked up the occasional Y-DNA I2 lineage in Anatolia and deeper testing of West Asians (particularly Armenians (https://www.familytreedna.com/groups/armeniadnaproject/about/results)) revealed Y-DNA I2 is present among them in non-trivial frequencies.

For the record, Nasidze et al.'s findings vis-a-vis Kurdish Y-DNA I have not been replicated yet. I should note quite a lot of the Nasidze papers focusing on West Asia have produced Y-DNA frequency results which, in retrospect, appear anomalous when compared to later and more expansive testing in the region. It's anyone's guess whether they were genuine hyper-drifted Y-DNA I, or the result of some sort of lab error.

I think its likely a result of the tiny sample sizes.

Motzart
07-07-2015, 04:50 PM
Kristina said:

"Just a thought: it may solve the oft-repeated question as to why uniparental markers in Sardinia (that is, I2a1) "don't match the autosomal DNA" (being EEF), to just imagine that I2a1-M26 was NOT necessarily in Europe during the WHG times in great numbers, but that the mass migrations were of the Cardial people, who had already picked up farming. In other words, that I2a1 is one of the few I2 clades that is not representative of WHG."


Yes it was. I2a1a ('pre-M26) already found in Mesolithic sweden , and Neolithic Paris basin, neither of which have anything to do with Cardial. ..

it isn't pre-m26, it is downstream from m26, I2a1a1a*

MT1976
07-07-2015, 11:33 PM
it isn't pre-m26, it is downstream from m26, I2a1a1a*

Ok. So all the more so...
I suspect M26 in Remedello - if correct- represents 'Neolithicized foragers' from the west, and not Cardial farmers from Greece

MT1976
07-09-2015, 01:56 AM
Although we need plenty more Mesolithic samples from Europe to tell for sure.