Thursday, September 22, 2016

The validity of Eastern Non-African as a clade based on genetic-drift

I got curious and asked David over at Eurogenes to try and run something to see how valid "Eastern Non-African" looked in terms of genetic drift shared between groups.


Green = various native populations have significant (>10%) "ENA" ancestry

For instance, would Han-Chinese people share more genetic drift with Onges and Papuans than with Villabruna-Cluster Hunter-gatherers (basically "WHG"), Ancient North Eurasians and Eastern European Hunter-Gatherers? Well, it does seem so:




Han-Chinese people share more drift with Onges and Papuans than they do with MA-1, the Karelia HG or the Villabruna HG which, I suppose, outright shows that they share more ancestry with the former two populations which implies their ancestors diverged from them later than when they diverged from the majority of the ancestry in VHGs, EHGs and ANEs.

One thing that intrigues me about this is how we apparently have very early evidence of "Modern Humans" in Oceania. From what I understand, there's archaeological evidence in places like Australia that imply Homo Sapiens Sapiens began inhabiting the region as early as ~50,000ybp which really makes things a little confusing if those Humans were the ancestors of modern Australian-Aborigines and, to some extent, also Papuans. [1] 

It's confusing because it's unlikely that the "clade" mostly ancestral to VHGs & ANEs differentiated from the "clade" mostly ancestral to ENAs that early, given that we have a Homo Sapien Sapien from that time-frame (Ust-Ishim, ~45,000ybp) whose genetic state predates this divergence:


Ust-Ishim is basal to all Out-of-Africa populations from East Asians to Onges to European & Siberian HGs and seemingly also Papuans and Australian-Aborigines [2]. The only exception so far being the theoretical "Basal Eurasians".

So, if the arrival date was ~50,000-40,000 years ago then ENAs, and whatever Paleolithic Europeans and Siberians are mainly descended from, would not, likely, have too notably diverged yet (or diverged at all) and it's thus incredibly unlikely that notable "ENA" substructure had formed by this point.



Basically, this implies modern Oceanians are either not the descendants of these early migrants at all or are perhaps some sort of mixture between them and later "ENA" migrants, with the latter group comprising much more of their ancestry? 

But I wonder if even the latter situation would cause modern Oceanians to look as "closer" to East Asians as they do. If they're a mixture between something ENA + a highly drifted Human population who may have not even participated in the traditional dispersals out of Africa + Denisovan admixture; they should look arguably much more divergent from other Out-of-Africa populations as a whole but Ust-Ishim looks to be basal to their non-Denisovan ancestry like he is to everyone else and they seem to share in a later ancestral clade with East Asians and the Andamanese that they do not share with Paleolithic Europeans and Siberians.

An abundance of ancient DNA from "Eastern Non-African" regions should make things more illuminating if anything, I guess. 


References:



2 comments:

  1. Given the results of the most recent papers, it looks like just Australians, from different sides of the same continent, have been extremely isolated from each other for tens of thousands of years.

    It also looks like there was maybe a tiny bit of gene flow among Papuans and ENA groups, and among Papuans and North Eastern Australians.

    To talk of Australians and Papuans as if they were one population is like saying that all of Eurasia and the Americas is just one population.

    ReplyDelete
    Replies
    1. "Given the results of the most recent papers, it looks like just Australians, from different sides of the same continent, have been extremely isolated from each other for tens of thousands of years.

      It also looks like there was maybe a tiny bit of gene flow among Papuans and ENA groups, and among Papuans and North Eastern Australians."

      I've got at least two of the new papers in my possession and skimmed through the one that made some interesting comments on Papuans (Genomic analyses inform on migration events during the peopling of Eurasia, Pagani et al.). It's pretty interesting and makes sense with what I've posted here. They basically find that Papuans mainly derive from the "Eurasian bottle-neck" but could derive just about ~2% of their ancestry from a Homo Sapien Sapien (AMH) population that hit-up regions outside of Africa before that which makes sense. If they trace any of their ancestry to very early OoA migrants who missed out on the "Eurasian-bottle-neck" and the supposed later splitting between the ancestors of ANEs+VHGs and ENAs; it would have to be a pretty small amount from what I can tell so something lower than 5% makes sense.

      "To talk of Australians and Papuans as if they were one population is like saying that all of Eurasia and the Americas is just one population."

      They are certainly distinct populations that have differentiated over-time. You're quite right in this respect. But, the point is simply that native Australians seem to relate to these ancients and moderns in a manner similar to how Papuans do (i.e. in that tree-mix) so that's why I was using Papuans as a stand-in for them. But if you'd prefer, here are the same stats with native Australians in them:

      Chimp Australian Mbuti Ust_Ishim 0.3324
      Chimp Andamanese_Onge Mbuti Ust_Ishim 0.3313
      Chimp Papuan Mbuti Ust_Ishim 0.3306
      Chimp MA1 Mbuti Ust_Ishim 0.3304
      Chimp Han Mbuti Ust_Ishim 0.3301
      Chimp Villabruna Mbuti Ust_Ishim 0.3282
      Chimp Karelia_HG Mbuti Ust_Ishim 0.3177
      Chimp Iran_Neolithic Mbuti Ust_Ishim 0.304
      Chimp Israel_Natufian Mbuti Ust_Ishim 0.3006

      Chimp Andamanese_Onge Mbuti Han 0.3863
      Chimp Australian Mbuti Han 0.3752
      Chimp Papuan Mbuti Han 0.375
      Chimp MA1 Mbuti Han 0.3547
      Chimp Karelia_HG Mbuti Han 0.3546
      Chimp Ust_Ishim Mbuti Han 0.3457
      Chimp Villabruna Mbuti Han 0.3389
      Chimp Iran_Neolithic Mbuti Han 0.3259
      Chimp Israel_Natufian Mbuti Han 0.3093

      Roughly the same results as Papuans. Ust-Ishim is still seemingly basal to native Australians' non-Denisovan ancestry and they clearly share more drift with the Han than ANEs, EHGs or VHGs do. So using Papuans as a proxy for them, in this case, was not misguided in my humble opinion.

      The problem still stands. Australian Aborigines don't make sense as 50,000ybp+ cut-offs from other Out-of-Africa populations with this sort of genetic profile. But things would make a little sense if they traced at least a bit of their ancestry to such people. Like that paper I mentioned suggested regarding Papuans...

      Delete