Saturday, July 4, 2015

Central Asians: A mixture between West Eurasians & East Asians

From looking at them alone many maybe curious about the genetic nature of groups like Kazakhs, Uyghurs, Uzbeks, Hazaras, Turkmen, Kyrgyz and so on. They can display extremely East Asian-esque features but in conjunction with this often display more typically West Eurasian phenotypic traits such as blue eyes or light brown to blonde hair.


The reason behind this is that at a basal level (ancient / pre-historic ancestry); they're a genetic mixture between West Eurasians and East Asians. The product of such intermixing that's been going on since the Bronze Age based on data from the recent Allentoft et al. 2015. [1]

Now, of course, Central Asians are no monolith. Many Tajiks, Uzbeks and Turkmen can honestly be by majority West Eurasian and perhaps only about ~20% East Asian (if not less).

Kazakhs who are in my experience far less heterogeneous can seem more like an even split between East Asians & West Eurasians but mostly all Central Asians do have some non-negligible East Asian ancestry and then also some non-negligible West Eurasian ancestry & are, for the most part, a cross between East Asians & West Eurasians.

You can see what I'm talking about above where Kazakhs, Turkmen, Uyghurs show a gross amount of West Eurasian (Early Neolithic Farmer, Western European Hunter-Gatherer & Ancient North Eurasian) ancestry in conjunction with East Eurasian (basically East Asian) ancestry. [note on South Eurasian]

And below you can see the heterogeneity I mentioned among groups like Uzbeks, Tajiks and Turkmen:

So I'd remain cautious with looking at the averages for various Central Asian populations run through the Eurogenes K8 Model (some info on the author of Eurogenes / David Wesolowski) based charts above and assuming virtually every member of each ethnic group would turn out like that.

Central Asia

This isn't just true for Central Asians though but for non-West Asian Turkic speakers at large (Kazakhs, Uzbeks, Turkmen, Kyrgyz & Uyghurs are all Turkic peoples). Altaians, Bashkirs & Yakuts who can be found in areas like Siberia or generally "Northern Eurasia/ Russia" (whilst being Turkic speakers) are also more or less "West Eurasian + East Asian":

This mixed nature is ultimately why Central Asians, various Siberians and such can display the kinds of sometimes unique looks that they do

This mixed nature is not just demonstrable via simple ADMIXTURE analyses and ancient DNA like with the recent Allentoft et al. but can be perceived often via PCAs (Principal Component Analysis)  as well where groups like Uyghurs & Yakuts often cluster as intermediates between West Eurasians and East Asians who more or less lack West Eurasian input (like the Japanese):

Central Asians like Kazakhs, Kyrgyz and so on, regardless of how they look, are not 100% East Eurasian by any means but are instead a cross been East Eurasians (East Asians in this case) and West Eurasians. Granted, many Central Asians and non-West Asian Turkic peoples can be more one than the other; with this sort of heterogeneity sometimes existing profusely in a single ethnic group like with Uzbeks.

I'm really no expert on Central Asia, Turkic speakers or really even Siberians but I was pseudo provoked into making a post like this upon encountering some people online who were bickering quite a bit over whether or not groups like Kazakhs were "purely" East Asian and well... This is the answer.

Reference List:


1. All of these groups do also carry Basal Eurasian ancestry as "ENF" (what Early European Farmer would be if you removed its WHG) basically caries Basal Eurasian.


  1. And now we know that the "Neolithic Farmer", "West Hunter-Gatherer", and "Ancient North Eurasian" components, likely originated from migrations from the Pontic-Caspian Steppe, correct?

    1. To some great extent, yes. But I'd say some of the West Eurasian-related ancestry in various Central Asians is also derived from other sources.

      From what I recall there's good evidence that a lot of the "Ancient North Eurasian" in areas like South Asia or Central Asia (though I'm less knowledgeable about Central Asia) most likely predates gene flow from the Pontic-Caspian steppe and there's also a good argument to be made for example that the peoples of cultures like the Indus Valley Civilization or the BMAC culture were rather rich in ancestries like "Neolithic Farmer".

      For example, the average Tadjik is sometimes more "Ancient North Eurasian" than the average Sintashta and groups like the Sintashta and Andronovo are what's ultimately relevant when it comes to the Steppe-derived ancestry in South and Central Asians given them seemingly being the early Indo-Iranian speakers.

      A random Afghan Tadjik for example being 31% "Ancient North Eurasian" can't entirely be explained by Pontic-Caspian Steppe ancestry. This Tadjik would need to be more than 100% Sintashta for this to be possible so it's quite clear that MA-1 related ancestry ("Ancient North Eurasian") and ancestries like "Neolithic Farmer" were likely to some extent present in areas of Central and South Asia before the arrival of pastoralists from the Steppe.

      We'll certainly no more someday with more ancient DNA from Central Asia.