June 30, 2009

Uyghurs as an admixed not source population (Xu et al. 2009)

This paper is interesting not so much because it estimates admixture in Uyghurs (click on the post label for previous studies on the topic), but because it explicitly rejects the hypothesis that they are a source ("donor") population.

If a population has substantial genetic variation which overlaps with that of two other groups, then there are two possible interpretations:
  1. It represents the population from which the other two groups sprang, or at least contributed genes to both of them
  2. It represents a mixture of the two other groups
What this paper does, is to show that Uyghurs are best explained as a mixture of Caucasoids and Mongoloids (#2) rather than #1.

Molecular Biology and Evolution, doi:10.1093/molbev/msp130

Haplotype Sharing Analysis Showing Uyghurs Are Unlikely Genetic Donors

Shuhua Xu et al.

Abstract

The Uyghur are a group of people primarily residing in Xinjiang of China which is geographically located in Central Asia, from where modern humans were presumably spread in all directions reaching Europe, east and northeast Asia about 40 kya. A recent study suggested that the Uyghur are ancestry donors of the East Asian gene pool. However, an alternative hypothesis, i.e. the Uyghur is an admixture population with both East Asian (EAS) and European (EUR) ancestries is also supported by our previous studies. To test the two competing hypotheses, here we conducted a haplotype sharing analysis based on empirical and simulated data of high density single nucleotide polymorphisms (SNPs). Our results showed that more than 95% of Uyghur (UIG) haplotypes could be found in either East Asian (EAS) or European (EUR) populations, which contradicts the expectation of the null models assuming that UIG are donors. Simulation studies further indicated that the proportion of UIG private haplotypes observed in empirical data is only expected in alternative models assuming that UIG is an admixture population. Interestingly, the estimated ancestry contribution of 44%:56% (EAS:EUR) based on haplotype sharing analysis is consistent with our previous estimation with STRUCTURE analysis. Although the history of Uyghurs could be complex, our method is explicit and conservative in rejecting the null hypothesis. We concluded that the gene pool of modern Uyghurs is more likely a sole recipient with contribution from both EAS and EUR.

Link

12 comments:

e-burke said...

What this paper does, is to show that Uyghurs are best explained as a mixture of Caucasoids and Mongoloids (#2) rather than #1.

Are there many who consider #1 a significant possibility.

Anonymous said...

Yes it is a rather peculiar idea, that the Uyghur population is anything other than an admixed population. Had the Uyghurs been an ancestoral population contributing to both the East Asian genepool and European genepool then their population today would contain a vast array of phenotypes from the European norm to the East Asian norm and many forms of intermediates. Instead they present a more or less uniform mix of phenotypes without the "extreme" European or East Asian looks.

I hate to say this, but the study is a waste of time and money.

Unknown said...

Who considers #1 to be a possibility? It seems redundant to do studies like this. While I believe race exists, there are usually no sharp, fixed lines between them. We often find continuums. From where the Uyghurs live, the farther west you go, the more caucasian the populations get, and the farther east you go, the more mongoloid.

Onur Dincer said...
This comment has been removed by the author.
Onur Dincer said...
This comment has been removed by the author.
terryt said...

"On one side, there are main clusters, some of which are very big".

What do you accept?

I'd claim at least five main clusters: Caucasoids and Mongoloids obviously make up two different clusters. I would also separate the Australoids into another cluster and divide Africa into two, West and an East/South combination. All other groups being a product of hybridisation, just like the Uyghurs. So how did these five very different clusters originate?

Dienekes said...

I don't think it's self-evident to people (although it is to me) that the Uyghurs are admixed. I recall that Spencer Wells documentary for example, who went to Central Asia to find the "missing marker" or whatever, and he started commenting on the mixed European/East Asian facial features of his host as if he was ancestral to both Europeans and East Asians.

Moreover, the analysis done here can be extended to other regions of the world (e.g., Africa) where historical/archaeological traditions are more ambiguous, i.e., where both alternatives are a priori valid as hypotheses.

pconroy said...

Dienekes,

I agree to both points. Yes, the interesting thing is to do the same analysis in Africa.

Terryt,

IMO, we will find that in Africa there are Pygmy (and maybe Khoisan), and Afrasians, with West Africans being a hybrid of the two.

pinkboi said...

Onur -

Race doesn't mean subspecies. It may have meant that to Coon, but we don't use that term in other species when referring to subspecies. I would define race to mean clusters. Accepting existence of race, but not subspecies is the norm today.

Everyone -

I don't think this study was a waste of time. Sometimes, looks can be deceiving and if they keep doing this on central asian populations, they may yet find people who are #1 or a mixture of #1 and #2...

terryt said...

"Accepting existence of race, but not subspecies is the norm today".

What's the difference between a 'race' and a 'subspecies'? The terms are used fairly interchangably for other species. It's only by convention that we don't use 'subspecies' when we refer to different human populations.

terryt said...

"IMO, we will find that in Africa there are Pygmy (and maybe Khoisan), and Afrasians, with West Africans being a hybrid of the two".

I've had time to consider your suggestion. So first off are Pygmies a single population, related to Khoisans, or are they the product of various groups of people who've been able to adapt to living in the dense jungle? If the latter they are not a main cluster. If the former I'm more than happy to include them as a main cluster. But they would easily fit into a South and East African cluster anyway.

Whether or not Pygmies and Khoisan are related it's widely accepted that any connection has been severed by a migration south and east of farming people originating in West Africa. The question arises: were these West African people actually the product of a hybrid swarm between two main clusters: Pygmies and Afraisians?

I can easily accept an expansion of Afrasian genes through the Sahel, possibly associated with cattle-raising. But did the Afrasian expansion take place through an Africa already occupied, and differentiated? In fact coastal people such as Ashanti, Ibo, Yoruba, and perhaps even Wollof, display a lesser influence of Afrasian genes. The question again arises: are these West Africans, and those who carried agriculture south and east, really the product of a hybrid swarm of Pygmies and Afraisians? If so they are not a main cluster. Otherwise they are.

Besides which, weren't the Afraisians simply the product of a hybrid swarm between Africans and Asians in the first place, and so not a main cluster?

Onur Dincer said...

What's the difference between a 'race' and a 'subspecies'? The terms are used fairly interchangably for other species. It's only by convention that we don't use 'subspecies' when we refer to different human populations.

Totally agree.