The Unz Review - Mobile
A Collection of Interesting, Important, and Controversial Perspectives Largely Excluded from the American Mainstream Media
 Gene Expression BlogTeasers
The Genetic Structure of East Asians
Email This Page to Someone

 Remember My Information



=>
Search Text Case Sensitive  Exact Words  Include Comments

There was a question about East Asian genetic structure. There have been a fair number of papers published on the issue. But over the years I’ve assembled a pretty large personal data set from public sources, as well as stuff people have sent me. I decided to look at the East Asian individuals and how they relate to each.

First, I focused on the major ethno-national groups (or ones of particular interest and relevance, such as Mongols). Second, I LD pruned the data set down to 96,000. Third, I did some outlier removal. For example, I wanted to include some Kalmyk data, but it turns out all the Kalmyk have European admixture at some level. And a subset of individuals from Cambodia and Vietnam are ethnic Chinese. Those were removed.

I ran ADMIXTURE K = 5 unsupervised. Treemix k = 500 and global rearrangements on, and rooted with Cambodians.

Eigenvalues 7.59041, 4.04862, 3.00559, 2.60692 and 2.01554.



Asian1

Asian2

Asian3

Asian4

AsianTreeix.9

AsianTreeix.8

AsianTreeix.7

AsianTreeix.6

AsianTreeix.5

AsianTreeix.4

AsianTreeix.3

AsianTreeix.2

AsianTreeix.1

Asian_htm_m40027816

First, the evidence of gene flow into the Han ethnicity, or the absorption of the Han of local substrate, is clear in these results. Second, I’m pretty sure that the weird affinity between Yakut (the northernmost Turks) and Cambodians has to do with admixture into both groups from non-East Eurasians that is old. In the case of Cambodians something Indian-like, and for Yakuts something more like “Ancient North Eurasians.”

 
• Category: Race/Ethnicity, Science • Tags: Asian Genetics, Genetics 
Commenters to Ignore...to FollowEndorsed Only
    []
  1. T. Greer says: • Website

    Do you know where the “Han South” and the “Han” were gathered from?

    Read More
    ReplyAgree/Disagree/Etc.
    AgreeDisagreeLOLTroll
    These buttons register your public Agreement, Disagreement, Troll, or LOL with the selected comment. They are ONLY available to recent, frequent commenters who have saved their Name+Email using the 'Remember My Information' checkbox, and may also ONLY be used once per hour.
    Sharing Comment via Twitter
    /gnxp/the-genetic-structure-of-east-asians/#comment-1010347
    More... This Commenter Display All Comments
  2. Frank says:

    Could you give a real quick (like 100 word) synopsis of what is known about the history of the Han from genetics?

    Something like, “They domesticated rice, so they could each have 20 children, similar to the people who domesticated wheat. But unlike the original wheat farmers, they somehow didn’t get overrun by people from the Steppe during the Bronze Age. This is why the Chinese are mostly lactose intolerant today.”

    Is that at all accurate?

    Read More
    • Replies: @Bill M
    I don't know if eastern steppe peoples like the Mongols and Huns were or are lactose tolerant. From what I understand they consume fermented dairy products.
  3. Bill M says:
    @Frank
    Could you give a real quick (like 100 word) synopsis of what is known about the history of the Han from genetics?

    Something like, "They domesticated rice, so they could each have 20 children, similar to the people who domesticated wheat. But unlike the original wheat farmers, they somehow didn't get overrun by people from the Steppe during the Bronze Age. This is why the Chinese are mostly lactose intolerant today."

    Is that at all accurate?

    I don’t know if eastern steppe peoples like the Mongols and Huns were or are lactose tolerant. From what I understand they consume fermented dairy products.

    Read More
  4. ELM says:

    An ever-present question involves the genetics of one specific Asian group: the Filipinos. Like most of Latin America, the Philippines was for a long time under Spanish rule. One school of thought goes that Filipinos all have some Spanish (or other European) ancestry. The other goes that only 3.6% of Filipinos have any Spanish ancestry whatsoever.

    The truth appears to be somewhere in between. The 3.6% figure refers to the percentage of Filipino males – according to one study – having a Y chromosome originating from Europe. This doesn’t, however, take account of Filipino men who might have Spanish ancestry on their mothers’ side. For example, an ex-boyfriend of mine who was born and raised in the Philippines but could ‘pass’ as a Southern Italian probably didn’t have a European Y chromosome because his Spanish ancestry came from his maternal grandfather’s side.

    However, Spain’s genetic impact on the Philippines was fairly minimal compared to that on Latin America, for instance. (In fact, I would say the example of the Philippines serves as a powerful counterweight to the notion that Latin America is really ‘Indian’ and not Western.) Most Spanish surnames in the Philippines don’t come from a European ancestor but rather from the fact that Filipinos were ‘assigned’ Spanish names by the authorities. (A few native surnames remain, plus a considerable number of names of Chinese origin.) So the Philippines sort of takes a different turn with respect to their genetic heritage.

    Read More
  5. Matt_ says:

    Good treemix graphics. Adding an African Outgroup like Khoi San might slightly change the visible structure, where rather than Yakut looking like a long branch within the Japanese-Korean clade, it would get its own branch.

    With the set of populations as it stands, I think, Yakut ends up being assumed by treemix to be extremely drifted within the Japanese-Korean clade because it has strongest affinity to JK while having high drift from all the other members of the East Asian group, who are all relatively closer together so get modelled as low drift. Adding an African outgroup would change that, I think, as treemix wouldn’t be able to place as high a level of drift on Yakut, as its drift from the African outgroup being relatively comparable to other East Asians would constrain giving it a higher drift overall. So it’s position would be solved by giving it a branch, and maybe an edge, I think, instead.

    Read More
  6. But where Cambodia get all these Yakut green from? That seems a lot! Right? Mongolised Turk invasion to India/South Asia, from there to Cambodia?

    Read More
    • Replies: @Shaikorth
    Razib is probably right that it's some kind of old shared ancestry represented that way. Whether it's ANE-related shared ancestry is another matter, for instance according to the f3-stats from Raghavan et al MA-1 isn't more related to Cambodians than to, say, Dai.

    Re: Yakuts, they aren't mongolized Turks, but more like turkified Tungusics (Fedorova et al. 2013 and Yunusbayev et al. 2015 both make it apparent that their closest relatives are Evenks) and their ethnogenesis isn't even a thousand years old. As a source of ancestry they aren't relevant outside of Yakutia.
  7. Shaikorth says:
    @PandaAtWar
    But where Cambodia get all these Yakut green from? That seems a lot! Right? Mongolised Turk invasion to India/South Asia, from there to Cambodia?

    Razib is probably right that it’s some kind of old shared ancestry represented that way. Whether it’s ANE-related shared ancestry is another matter, for instance according to the f3-stats from Raghavan et al MA-1 isn’t more related to Cambodians than to, say, Dai.

    Re: Yakuts, they aren’t mongolized Turks, but more like turkified Tungusics (Fedorova et al. 2013 and Yunusbayev et al. 2015 both make it apparent that their closest relatives are Evenks) and their ethnogenesis isn’t even a thousand years old. As a source of ancestry they aren’t relevant outside of Yakutia.

    Read More
  8. terryt says:

    “First, the evidence of gene flow into the Han ethnicity, or the absorption of the Han of local substrate, is clear in these results”.

    That makes complete sense. It is pretty obvious that the Mongoloid phenotype did not originate in SE Asia but much further north. It is relatively recent into SE Asia as early humans there are considered ‘Papuan-like’. The change occurs in the Neolithic, presumably with a movement south from where the eastern Neolithic originated in northern China. The Han would be just the most recent of these movements. It would be very revealing to be able to include Papuans and Australian Aborigines to give us a better idea of what was happening. Perhaps it is represented by the yellow in Filipinos.

    Read More
  9. @Bill M
    I don't know if eastern steppe peoples like the Mongols and Huns were or are lactose tolerant. From what I understand they consume fermented dairy products.

    They’re not.

    Read More
  10. The ethnogenesis of what we now call East Asians really interests me, because it’s one of the few large missing data points still remaining. We know Amerindians, Europeans, South Asians, Southeast Asians, and most Sub-Saharan Africans are genetic composites. The only areas we don’t have similar indications yet are East Asia, “pure” Austro-Melanesians, and the Near East.

    We know from archaeological remains that Austro-Melanesians lived at least as far north as the Red River valley in Southern China. IIRC the relatively minor differences between the genetic background of Southeast Asians and East Asians have to do with the varying levels of admixture from this component. What I find myself asking is if some similar populations lived as far north as the Yellow River – if East Asians were, similar to other agriculturalists, a hybrid group – with (non Ancient North Eurasian) Siberians and some other, now extinct group being the parent populations.

    Of course, maybe East Asians are exceptions to the rule – the one hunter-gatherer population which was not subject to significant admixture in the Mesolithic and Neolithic. There’s no reason to think complex admixture had to be universal. Time will tell, since I think Razib has intimated Reichlabs is on the case.

    Read More
    • Replies: @Shaikorth
    With some tweaking of ADMIXTURE, East Asians can be represented as even three-way mixes though whether it's meaningful is a different story.
    http://oi60.tinypic.com/f4pquu.jpg
    The "African" if it's anything real might be some kind of basal stuff making them distinct.

    Graph from here: http://www.nature.com/ncomms/2015/150324/ncomms7596/full/ncomms7596.html

    Agree that what knowledge we have now isn't enough to make a major conclusion about this.
  11. Shaikorth says:
    @Karl Zimmerman
    The ethnogenesis of what we now call East Asians really interests me, because it's one of the few large missing data points still remaining. We know Amerindians, Europeans, South Asians, Southeast Asians, and most Sub-Saharan Africans are genetic composites. The only areas we don't have similar indications yet are East Asia, "pure" Austro-Melanesians, and the Near East.

    We know from archaeological remains that Austro-Melanesians lived at least as far north as the Red River valley in Southern China. IIRC the relatively minor differences between the genetic background of Southeast Asians and East Asians have to do with the varying levels of admixture from this component. What I find myself asking is if some similar populations lived as far north as the Yellow River - if East Asians were, similar to other agriculturalists, a hybrid group - with (non Ancient North Eurasian) Siberians and some other, now extinct group being the parent populations.

    Of course, maybe East Asians are exceptions to the rule - the one hunter-gatherer population which was not subject to significant admixture in the Mesolithic and Neolithic. There's no reason to think complex admixture had to be universal. Time will tell, since I think Razib has intimated Reichlabs is on the case.

    With some tweaking of ADMIXTURE, East Asians can be represented as even three-way mixes though whether it’s meaningful is a different story.

    The “African” if it’s anything real might be some kind of basal stuff making them distinct.

    Graph from here: http://www.nature.com/ncomms/2015/150324/ncomms7596/full/ncomms7596.html

    Agree that what knowledge we have now isn’t enough to make a major conclusion about this.

    Read More
  12. YS says:

    Just a random but somewhat related question. I’m East Asian and I was wondering what everyone’s thoughts are on 23andme. Should I use their or another personal genomics service? I have privacy concerns but not of the “I’m a special snowflake” variety but of the aggregation of our entire identities in the databases of large companies variety. I think it’s somewhat obscene that Google, with 23andme’s data in hand, could essentially know everything about you. Isn’t that kind of messed up? And they also profit financially from ownership of this information. I know most of you think these fears are vain and unfounded but I would still like to get a nudge in the right direction before deciding. I am curious about my ancestral background and other genetic information though I don’t know if I’m curious enough.

    Read More
    • Replies: @Frank
    I wouldn't worry too much about DNA privacy. It just isn't really worth that much yet.

    If you don't want only 23andme to have it, you can give it out freely. Many people are doing that on the web.

    You can also use a fake name when you send it in.
  13. Frank says:
    @YS
    Just a random but somewhat related question. I'm East Asian and I was wondering what everyone's thoughts are on 23andme. Should I use their or another personal genomics service? I have privacy concerns but not of the "I'm a special snowflake" variety but of the aggregation of our entire identities in the databases of large companies variety. I think it's somewhat obscene that Google, with 23andme's data in hand, could essentially know everything about you. Isn't that kind of messed up? And they also profit financially from ownership of this information. I know most of you think these fears are vain and unfounded but I would still like to get a nudge in the right direction before deciding. I am curious about my ancestral background and other genetic information though I don't know if I'm curious enough.

    I wouldn’t worry too much about DNA privacy. It just isn’t really worth that much yet.

    If you don’t want only 23andme to have it, you can give it out freely. Many people are doing that on the web.

    You can also use a fake name when you send it in.

    Read More

Comments are closed.

Subscribe to This Comment Thread via RSS Subscribe to All Razib Khan Comments via RSS
PastClassics
Confederate Flag Day, State Capitol, Raleigh, N.C. -- March 3, 2007
The major media overlooked Communist spies and Madoff’s fraud. What are they missing today?
Are elite university admissions based on meritocracy and diversity as claimed?
The “war hero” candidate buried information about POWs left behind in Vietnam.
The evidence is clear — but often ignored