search this blog

Friday, October 30, 2015

European-specific mtDNA U5a2a in ancient remains from Xinjiang, northwest China

I'm skeptical that a 700-year-old mito genome can be used as evidence of prehistoric population movements. Obviously, for that sort of thing we'd need a sequence from before recorded history. Nevertheless, the discovery of mtDNA haplogroup U5a2a in the Eastern Pamirs is still very interesting.

The complete mitochondrial genome of one 700-year-old individual found in Tashkurgan, Xinjiang was target enriched and sequenced in order to shed light on the population history of Tashkurgan and determine the phylogenetic relationship of haplogroup U5a. The ancient sample was assigned to a subclade of haplogroup U5a2a1, which is defined by two rare and stable transversions at 16114A and 13928C. Phylogenetic analysis shows a distribution pattern for U5a2a that is indicative of an origin in the Volga–Ural region and exhibits a clear eastward geographical expansion that correlates with the pastoral culture also entering the Eurasian steppe. The haplogroup U5a2a present in the ancient Tashkurgan individual reveals prehistoric migration in the East Pamir by pastoralists. This study shows that studying an ancient mitochondrial genome is a useful approach for studying the evolutionary process and population history of Eastern Pamir.

Ning et al., Ancient mitochondrial genome reveals trace of prehistoric migration in the east Pamir by pastoralists, Journal of Human Genetics advance online publication 29 October 2015; doi: 10.1038/jhg.2015.128


Maju said...

It is consistent with previous data from other parts of Central Asia that indicate that the West/East divide, very clear in the Bronze Age, was vanishing in the Iron Age. Intriguingly it did not just imply migration of Western lineages to the East but also of Eastern lineages to the West.

Alberto said...

Yes, by the hints we have from that Scythian genome from Mathieson et al. it seems that the movements were back and forth along the steppe between Eastern Europe and S-C Asia. While Sintashta/Andronovo were clearly more European than Yamnaya/Afanasievo, the Scythian sample looks again more Asian than even Yamnaya/Afanasievo (pending further analysis). Probably East Asian started to play a role by then too.

Davidski said...

That Scythian has some Siberian ancestry, probably from around the Altai, but I doubt he has any South Central Asian admixture.

The idea of a back and forth movement between Eastern Europe and South Central Asia doesn't make much sense because of the lack of ASI in Eastern Europe. ASI is present in Iran, the Hindu Kush, and even the Pamirs, and it's difficult to argue that it only got there very recently.

There are isolated cases of ASI admixture in the Balkans, but these can be explained by Roma ancestry, so they're not related to population movements across the steppe.

Alberto said...

That Scythian genome seems to plot south of Yamnaya types (not by much), and admixture shows the highest "Armenian" and least Euro_HG, plus something probably Siberian. We have to wait, but he's clearly not going to be a Sintashta/Andronovo type. Maybe a mix of that with "Armenian", which is what suggests movement back and forth with a good amount of mixing.

The question of ASI is interesting, but difficult at this point. I do think that it's arrival as far north as Tajikistan and Turkmenistan has to be from those BA/IA movements. I doubt that ASI was in that area before the Bronze Age. Modern East Europeans don't need to have any ASI (which was diluting fast as you get further away from India), especially when they don't even have Scythian admixture (hardly any R1a-Z93). The final movement in the steppe was west to east and modern Russians are basically North Europeans with a bit of Siberian admixture.

Why are there no news about ancient DNA from BMAC and earlier (Jeitun) coming? Is no one interested in sampling those areas? Maybe political difficulties? Same for Iran.

Davidski said...

Scythians came from the Altai/South Siberia region, and were active in the Kuban steppe, hence the Siberian and inflated Caucasus admixture in this individual.

Really not seeing any connection to South Central Asia here.

Alberto said...

Inflated Caucasus admixture from South Siberia? Well, if those "Armenians" were actually from South Siberia then it does makes sense.

Let's wait to see the genome first. Maybe it indeed looks like a Andronovo + Siberian type. But if it does have more "Armenian" than Yamnaya as it seems (and much more than Sintashta/Andronovo, who were the people in that region before them), I don't know why you don't want to see a southern influence in it.

Alberto said...

Ah, sorry, you meant the inflated Caucasus admixture from the Kuban steppe. Yes, that could be too. It will be difficult to know with just Scythian samples where the "Caucasus" admixture came from exactly, except that from the south. We'll need sampling from all the southern areas of the steppe to be able to tell there difference, if any.

Davidski said...

The Caucasus admixture probably came from the Caucasus, which is close to the Kuban steppe.

capra internetensis said...


There is already a Bronze Age sample from Turkmenistan, but it is so degraded as to be useless. I have no doubt people are very interested, though politics may be a problem as you say.

Rob said...


I might have asked before, but what processes what you put down for the putative central Asian movement to the steppe between 5000-
3500 BC ?

I know the south -east caspian region had early Agro-pastoralism and evidnec of copper experimentation, but in 4000 BC it appears it was a case of 'steady as she goes' there, rather than anything revolutionary - as in the Caucasus region. Certainly, no real evidence of out-migration, technological revolution or state-society formation .

Alberto said...


You are right, nothing too revolutionary was happening in that area around (or before) 5000 BC when the first migrations might have happened. Samara/Khvalynsk cultures themselves are noting revolutionary compared to any other Eneolithic culture, though being Eneolithic cultures in itself was quite revolutionary compared to the Mesolithic societies that preceded them.

The point is that Mesolithic hunter-gatherers can't suddenly develop an Eneolithic culture on their own, in a few centuries. This is only possible if the culture is brought by a new population who has already developed it, even if it's not revolutionary compared to any other one of the time (or even behind, depending to what you compare to).

Ancient DNA already has confirmed this: the "revolution" (from Mesolithic to Eneolithic) arrived with a new population (it couldn't have happened otherwise).

But you really don't need that he population that arrived came from a place were revolutionary things were happening. Probably they just moved because of environmental causes. For example, the site Ayakagitma in Uzbekistan, that I've mentioned a few times before, had domestic animals since around 6000 BC (cattle, sheep/goats, camels, maybe horses?). In the layer from 5700-5500 BC. the horse remains (domestic or not) were 43.6%, while camel remains were 12.9%. Then in the next layer, from 5500-5400 BC. horse remains go down to 10.7% and camel remains up to 50%. And after this the site is abandoned for 1400 years (and when it's occupied again, camel remains rise to 84.5%. Today is an absolute desert). Why did these people abandon the site? And where did they go? One option is north, to the steppe.

Innovations, from SE Iran to Turkmenistan and reaching the Caucasus happened a bit later. But I don't see why we would need these innovation to justify an early migration. The migration did happen, in any case: we have the samples that prove it. From where exactly is difficult to say, so the reasons are therefor unknown too. But the options are not that many, and given the genetic profile of this "Georgian-like" population it fits with the places from where it could enter the steppe (except maybe with Tripolye, which we presume to be more "Anatolian-like").

The Caucasus itself is a bit of a mystery. But given how close Transcaucasia and Anatolia are, and given that early Transcaucasian cultures are clearly related to those of the Near East, it seems unlikely that they were genetically "Georgian-like" (we would have seen ANE in EEFs, I think). So if these new people arrived to Transcaucasia in the late 4th millennium (and to the North Caucasus already with Maykop), that's already a bit late to be the source area of the population that arrived to Samara/Khvalynsk earlier than that. Though perfectly in time to have a cultural impact, regardless of genetic impact or not.

The other option, of course, is what David thinks: Mesolithic hunter-gatherers at some point started to take southern wives and somehow abandoned their lifestyle and culture for that of their wives (or simply developed an Eneolithic culture on their own, and the wives are just a coincidence in time and lifestyle/culture).

Davidski said...

There's really no evidence of a new population arriving on the steppe during the Eneolithic.

So far we only have evidence of gradual gene flow from the Caucasus exclusively via maternal admixture.

If you want to see what a sudden population turnover looks like, then the best example so far is Late Neolithic Central Europe.

Rob said...


Fair enough; certainly plausible hypothesis


Well I think it's relative and samples from Poland will nail down the pace of change in Europe.

But there are some solid suggestions for movements onto the north Caucasus and steppe; but cannot refute or confirm given the present state of aDNA data .

Whatever the case, movement into the steppe probably did occur from neighbouring regions; like western forest steppe and north Caucasus

Krefter said...

"So far we only have evidence of gradual gene flow from the Caucasus exclusively via maternal admixture."

Or it could be there was a lot of paternal admixture that was replaced by two mega founder effects. 99% of EHG Y DNA died out. Two EHG lineages dominated.

Grey said...

"The point is that Mesolithic hunter-gatherers can't suddenly develop an Eneolithic culture on their own, in a few centuries. This is only possible if the culture is brought by a new population who has already developed it, even if it's not revolutionary compared to any other one of the time (or even behind, depending to what you compare to)."


I'd have thought hunter-gatherers living in a region with copper on the surface might be able to create cold worked copper beads for decoration without needing to develop a neolithic culture to go with it.

So it seems to me you could have the two halves of an Eneolithic culture living adjacent to each other: neolithic farmers without copper and mesolithic hunter-gatherers with copper.

Grey said...

...leading to a trade of copper beads for other stuff - like women - and an eventual hybrid culture.

Rob said...


Further : but yes I think you're correct that it's likely the switch from Mesolithic to Eneolithic entailed new population arriving

Davidski said...

What evidence do you have of a new population arriving on the steppe during the Eneolithic?

Intermarriage between steppe men and Caucasian women doesn't count.

katja-kristiina said...

According to Bermisheva et al paper on Volga-Ural mtDNA, that haplotype with mutation 114A is typical of Tatars but it is also found in Bashkirs and Komi-Permjaks and Mordvins. In the west, it is quite frequent in Finland and the frequency of U5a2a in Belorusia is 1.12% (while U5a1 is at 2,25%), but it seems to be rare in Scandinavia.

Mutation 114A was present already in Mesolithic Germany as it is found in Hohlenstein-Stadel 6700 BC (U5a2a) and Late Eneolithic Bulgaria in Durankulak 5500-4000 y. a (U5a2). It has also been found in Vatya, Corded Ware and Unetice burials.

katja-kristiina said...

It looks like U5a2 is of Central European origin as Neolithic Dnieper Donets and Dniestr and Donau Yamnaya as well asCatacomb cultures have yielded U5a1.

Rob said...


There is suggestive evidence, no doubt. For example, evidence of movement from Cucuteni-Tripolye onto the lower Danube -Dniester - Bug interfluvial. Similarly, there is evidence of movement from the north Caucasus (ie Majkop) onto the lower Don stepp, viz the Kostantinovka group. These begin to appear c. 3700 - 3500 BC. I know Khvalysk dated from 4200 already had Teal, but the dated interval of the samples are actually very broad (4200-3500 BC). Either way, the most Teal arrived after 3800 BC. It is also in these sub-regions where the earliest yamanaya dates come from.

There were several regional groupings in the greater Pontic zone, stretching from the Danube to the Carpathians to the Caucasus and perhaps beyond, with evidence of movements within this zone. I do not make any conclusions as to how this translates into R1a and R1b specifically, as we need considerably more samples than that.

Alberto said...


"What evidence do you have of a new population arriving on the steppe during the Eneolithic?

Intermarriage between steppe men and Caucasian women doesn't count."

The Khavynsk genomes don't give a clear enough picture. They are only 3, and one of them is doubtful to even be part of the culture, he looks like some outsider who was killed. One of them seems to cluster with Yamnaya already, while other is intermediate between Yamnaya and EHGs. More samples are needed to clarify the details of the situation, but it's already clear that new people arrived (men or women, or both).

Genetically, your idea of intermarriage based on EHGs having old clades of R1a and R1b is parsimonious. But anthropologically it isn't. You have to explain somehow the leap from Mesolithic to Eneolithic culture, and the only possible explanation is outside influence (which now we know that was almost in every case demic and not simply cultural, and in this case we have genomes that show it too).

The only thing you're hanging onto is that R1a and R1b were exclusively on the steppe with EHGs and therefor the continuity is proved by male lineages. If R1a/b appear out of the steppe I'm afraid you won't have a case. (In your defense, I must say that no one else has stepped up to propose any better hypothesis to defend the continuity between Mesolithic EHGs and Eneolithic steppe cultures. Unless we count persons like "a" showing admixture results of Kostenki14 and Yamnaya to prove the continuity and similar no-cases).

Davidski said...

I'm sensing some misunderstandings here about the EHG people.

Not all Mesolithic populations were the same, and EHG groups were actually quite advanced. For instance, they had pottery, which they probably inherited from their Western Siberian ancestors.

Alberto said...

I did post some paper about the possible origins of the pottery a while back. It's not clear, as usual, due to the scarce archaeological data, but it looks like it arrived from the East Caspian more than from Siberia (as probably the ancestors themselves). The Mesolithic hunter-gatherers from the South Caspian (and around Iran, but there are few details about the latter) were indeed rather advanced, with even some evidence of attempts of animal domestication, though that finally came with people from the Near East.

But still, from Mesolithic (even with some pottery) to the Eneolithic PIE cultures there's quite a long way and not enough time. And the arrival of new people with ENF is already proved. So anthropologically it's still way more parsimonious that the culture came with the new people.

Grey said...

Maybe worth pointing out that nuggets of copper, like gold can be found in rivers in source regions so can be picked up off the ground - no smelting - and the earliest form of metallurgy is just warming in the fire to make it softer and then hitting it with a rock to make simple shapes like beads.


Damara people of Namibia - classified as HGs at the time but some of them living in settlements making copper jewelry for trade.

capra internetensis said...

I don't buy that teal came from EHG guys going a thousand miles from Samara to the Caucasus to pick up girls.

The "new population on the steppe" being a pastoralist fusion of EHG locals and teal farmers (from the southern steppe/Caucasus piedmont region) to begin with would make sense. The sex bias in uniparental markers (granting it is real) can be the result of patrilocality during the formation of this hybrid population (some selection is also possible). The eventual dominance of R1a-M417 and R1b-M269 specifically is chance founder effect.

It is also still possible that Alberto is right and that the "new population" is a mix of (Kel'teminar?) EHG and South Central Asian teal farmers from further east. We need more aDNA before we can distinguish between the many possible scenarios.

Davidski said...

You don't have to buy that teal came from EHG guys going a thousand miles from Samara to the Caucasus to pick up girls, because no one ever claimed that this is what happened.

First of all, the women didn't need to be picked up, they were highly mobile in patriarchal societies during the Bronze Age. Secondly, a lot of the initial admixture into the more mobile steppe groups probably happened just north of the Caucasus.

It is also still possible that Alberto is right and that the "new population" is a mix of (Kel'teminar?) EHG and South Central Asian teal farmers from further east.

Lots of things are possible, but there's not much point going on about things that are highly improbable.

capra internetensis said...


Okay, I don't think we really disagree then. A hybrid population formed near the Caucasus is plausible. No need for exclusive long-range bride exchange, just regular interaction with a reasonable level of sex bias.

Matt said...

I think it makes sense to assume the approximately 50% Samara ancestry in Yamnaya on present evidence.

Ultimately, at the moment we do not know the ultimate geographical extent of EHG ancestry cultures though, so it's possible that this extended south of the steppe to the East of the Caspian and that this is where much of the real admixture actually happened (just as much of the WHG ancestry in Middle Neolithic Europeans is just from Anatolia EN). IRC cranial type of Kelteminar Culture suggests similar ancestry to EHG? That culture has some association with stockbreeding, but this is not thought to be much linked to Yamnaya (but perhaps evidence will be found...). No way to be sure right now, so best just to assume the most simple model and keep an eye to testing others in the long term.

Grey said...

Two adjacent populations + bride exchange + patrilocality + bigger effect on smaller population is probably the simplest explanation

but look at the map.

Samara is an obvious place for a river trading post.

So if the people who lived there had something valuable to trade down the river to the two seas then I don't think long range bride exchange as a result is strange - only the scale of the local founder effect.