Saturday, October 10, 2015

All Modern Africans are part Eurasian? We should probably be a bit cautious

In light of a recent study that sampled a 4,500 year old Southwestern Ethiopian (labeled "Mota" for the cave he was found in) we now have data suggesting that Africans all around other than the obvious already known cases like Horn Africans & various Southeast Africans; supposedly have non-negligible Eurasian ancestry.

Truthfully, upon really taking a look at this paper's various results... I must personally advise some caution and say that I'm not 100% convinced that this Eurasian ancestry in all these groups is real. Why? Well, it's mostly but not entirely the scope as well as the levels. 

Every population from Hunter-Gatherer Pygmies like Mbutis & Biakas [note] down to farmer groups like the Yoruba or pastoralists like Anuaks are all supposedly about ~6-8% Eurasian (supposedly even West Eurasian?)? What's suspicious about this is the general homogeneous amount found all across in all these populations that prior to this study showed no strong Eurasian affinities other than some signs of Neanderthal ancestry or a hint here and there in some admixture runs in only some of them, if I recall correctly. [2]

But what large scale event occurred that spread Eurasian ancestry at such a homogeneous level to so many culturally & genetically distinct as well as geographically far removed from one another populations? This doesn't entirely even make sense. 

It can make some sense when observing some populations like Yorubas showing prior signs of Neanderthal ancestry and then the presence of markers like mtDNA U in West Africa but all across Africa? From Nilo-Saharan speakers like the South Sudanese ("Sudanese" samples) to Niger-Congo speakers like South African Bantu speakers and at a roughly equal level?

I mean there's nothing shocking about Eurasian ancestry showing up in groups like the Nama whom we know have both South Cushitic pastoralist & European admixture or in the Maasai who are substantially South Cushitic (and perhaps even part East Cushitic?) admixed or in Horn Africans but in these other populations it's surprising but would to my mind be more acceptable if these levels varied more.

F.e. if the South Sudanese were like ~3% and Yorubas were like ~7% & things just varied altogether then these results would seem a bit more plausible to me as it would mean there was some very ancient Eurasian expansions into Africa (plenty of evidence for such) that just so happened to hit these populations in different ways and likely at different times and then understandably resulted in different levels of Eurasian admixture.

But then Mbutis are ~6%, Yorubas are ~7% , Southern African Bantus are ~7% & the South Sudanese are ~6.5%... I must ask how? Why such extremely close levels? I'm not going to come out and say these results are 100% moot but I must say that one should be a little suspicious of them.

Perhaps this is due to Mota containing some form of a very archaic African element that all these populations lack? [note] This could perhaps shift them all away from Mota and towards Eurasians at a roughly equal rate? I'm really just shooting in the dark here as I'm generally just in shock over this study's results.

All I can say at this point is that further study is required... More ancient genomes from Africa, more from Eurasia and I'll be excited to see what we get when people like David who runs Eurogenes or Kurd over at Anthrogenica get their hands on this sample and start comparing him more and more to modern populations.

A lot more still needs to be learned and tested but for now this study's claim that these groups have Eurasian admixture is what the data we currently have is saying...

Reference List:

1. Ancient Ethiopian genome reveals extensive Eurasian admixture throughout the African continent, Llorente et al. (the study's text is pay-walled but the supplementary information which contains most of what you'll need anyway is free)



  1. I do not think you should be so skeptical about this. All Africans have some Eurasian "back-flow". But all Eurasians are just a branch of Africans (with a bit of Neanderthal and/or Denisovan, which probably also arose from Africa at an earlier time). In the long run we all come from the same place.

    I think this Mota genome is very enlightening, and I can't wait until there are even older African genomes. I suspect that most Africans have much more "Eurasian" ancestry than just this recent EEF-like ancestry. This is impossible to tell without more ancient genomes.

    If that is true, then all 'out-of-Africa' populations will turn out to be even more Neanderthal than can be predicted right now.

    The human family tree just keeps getting more interesting all the time.

    1. Indeed, the tree does keep getting more interesting with the more ancient genomes we have. I'm not entirely against the notion that various African populations having Eurasian ancestry to some extent; there is prior proof for this among some such as various West Africans (f.e. Yorubas) showing dabs of Neanderthal ancestry for example (then there's mtDNA U6 and so on).

      I'm honestly just waiting for more data at this point. More ancient genomes from various parts of Africa, more Eurasian ancient genomes and then we can be sure but for now the equal jump all these distinct populations are showing in Eurasian ancestry is just a little suspicious in my humble opinion.

      But thanks for the comment, man. People like you are exactly the types I like to attract with this blog. Sorry for taking so long to approve your comment btw, I had a lot of work due at my university over the last week and was off the grid for days.