Page 2 of 2 FirstFirst 12
Results 11 to 16 of 16

Thread: Admixture History of Uyghurs

  1. #11
    Registered Users
    Posts
    513
    Sex

    The hypothetical population groups used are far too large and the resolution is not high enough. For example, if you look at the ADMIXTURE estimates, the Uyghur have the same quantity of "South Asian" admixture as Anatolian Turks (~17%) and even Bedouins and Jordanians have ~12% "South Asian" admixture. The population with the largest quantity of "South Asian" admixture outside of SC Asia are Lezgins and Kumyks at about a quarter, and these two populations have, quite possibly, the least recent population contact with South Asians among all modern West Eurasians.

    The fundamental problem here is that all West Eurasians are forced into K=2 in this analysis and become admixtures of "Basques" and "South Eurasians". Under such an admixture model, its not quite clear what the "South Asian" admixture is representing, whether in HAPMIX or in the ADMIXTURE history graph. Especially when the most "South Eurasian" Europeans become Cypriots and Sicilians here, and also when Mala are ~100% "South Asian" but then Kalash are ~80% "South Asian". Are the Kalash ~80% Mala ancestry? Really not a realistic representation here.

    Presumably you can run a HAPMIX for the "South Asian" admixed Middle Easterners as well but the South-Asian like segments in Cypriots and Sicilians map onto Iran_N and CHG instead of ASI in Mala. It would be much improved if admixture history was investigated at K=4 in West Asia instead of K=2, in fact K=6 in West Eurasia would be even better, and Pugach et al were able to do this with Turkic populations using sources that would be very similar to each other (West Siberian, Central Siberian, North East European etc), so I don't see why they wanted West Eurasians to be split into 2 parts only. At higher K we can distinguish between Mala-like ancestry and something centred in West Asia.

    Quite inclined to say that the Uyghur in fact have very small quantities of South Asian ancestry instead of being 17% "Mala" when Anatolian Turks are obviously not 17% contribution from South Asians proper.
    Last edited by Ryukendo; 06-18-2017 at 12:45 AM.

  2. The Following 4 Users Say Thank You to Ryukendo For This Useful Post:

     DMXX (06-18-2017), Michał (06-18-2017), rozenfeld (06-18-2017), Tsakhur (06-18-2017)

  3. #12
    Registered Users
    Posts
    613
    Sex
    Ethnicity
    Parsi Brahmin
    Y-DNA
    L-SK1414
    mtDNA
    U8b1a1

    African Union Ainu AchaemenidEmpire1 Kurdistan Star of David Dravida Nadu
    Interesting. There are some uyghurs on Gedmatch and they're often modeled as half Mongolian half Pamiri Tajik(at fairly
    Close distances). Wouldn't be surprised if the pre Turkic population of the Tarim basin was pretty similar to modern Pamiris.

  4. The Following User Says Thank You to jesus For This Useful Post:

     Awale (06-18-2017)

  5. #13
    Registered Users
    Posts
    564
    Sex
    Y-DNA
    C-M217
    mtDNA
    M8a

    Kyrgyzstan
    Quote Originally Posted by jesus View Post
    Interesting. There are some uyghurs on Gedmatch and they're often modeled as half Mongolian half Pamiri Tajik(at fairly
    Close distances). Wouldn't be surprised if the pre Turkic population of the Tarim basin was pretty similar to modern Pamiris.
    Actually, there are Pamiris living in Tarim basin: https://en.wikipedia.org/wiki/Tajiks_of_Xinjiang (they are called Tajiks, but actually they are Pamiris)

  6. The Following 3 Users Say Thank You to rozenfeld For This Useful Post:

     Awale (06-18-2017), Michał (06-18-2017), Tsakhur (06-18-2017)

  7. #14
    Registered Users
    Posts
    30
    Sex

    United States of America San Marino Italy Veneto Italy Piemonte Occitania
    Quote Originally Posted by Ryukendo View Post
    The hypothetical population groups used are far too large and the resolution is not high enough. For example, if you look at the ADMIXTURE estimates, the Uyghur have the same quantity of "South Asian" admixture as Anatolian Turks (~17%) and even Bedouins and Jordanians have ~12% "South Asian" admixture. The population with the largest quantity of "South Asian" admixture outside of SC Asia are Lezgins and Kumyks at about a quarter, and these two populations have, quite possibly, the least recent population contact with South Asians among all modern West Eurasians.
    Can you please post their Admixture results here? I don't have full access to the paper unfortunatey.

  8. #15
    Administrator
    Posts
    2,322
    Sex
    Y-DNA
    R2a*-M124 (L295-)
    mtDNA
    D4j5*

    England
    Quote Originally Posted by Immchr View Post
    The Indo-Aryan Prakrit influence dates to, at the earliest, around 2500 YBP.
    You are correct, sir.

    But the paper argues that the WE-SA admixed population entered the Tarim Basin as a single population and then got admixed with a East Asian - Siberian group. This admixture with the EA - SIB group itself is dated to 3750 YBP i.e. 1750 BC. This is also most likely an underestimation as the authors argue. The more probable date would be around 2000 BC. In other words, the West Eurasian and South Asian admixed group entered the Tarim Basin before 2000 BC. How is the Prakritic influence of later period of any relevance here ? Can you explain ?
    Per my earlier comment:

    Their admixture models don't mean much without West, South and Central Asian aDNA from the Eneolithic onwards.
    Their admixture model is theoretical. You're using it as evidence against the P-C steppe theory in the absence of aDNA from the region. I don't need to argue anything beyond that for obvious reasons.

    One problem with the Tarim is that archaeological surveys of the region have been pretty scarce since the era of Aurel Stein (when the Tocharian texts were first discovered). There's evidence of Mesolithic culture overlap between the region's HG's with southern Siberia and even Afghanistan. Yet, we don't have a full understanding of these people, and their potential movements (microliths aren't sufficient for inferences).

    There well could have been a currently unexplained influence of South Asian heritage in the Tarim that predated the Indo-Aryan cultural influence there. I'm not opposed to the idea. Just to the utilisation of this study as evidence in favour of that scenario.

    And most importantly, the authors argue (please go through the OP once again), that the West Eurasian ancestry, across all the Uyghur groups in the Tarim Basin, positively correlates with the South Asian ancestry. This is only possible if the West Eurasian ancestry and South Asian ancestry entered together as a single admixed group in to the Tarim Basin.

    Further, this WE-SA admixture is higher in the Southwest and declines as one goes Northeast. In other words, the WE-SA population most likely entered the Tarim from the Southwest i.e. from Southern Central Asia. If there was a separate West Eurasian intrusion from the steppe, then the West Eurasian ancestry in Tarim would not positively correlate with the South Asian ancestry. The pattern of West Eurasian admixture should have no relation to the pattern of South Asian admixture. But that is not what we find. There is not a single group of Uyghurs which only have the West Eurasian ancestry without the South Asian admixture.
    Correlation between components derived from modern populations isn't indicative of much. By that reasoning, Y-DNA J2 and R1a1a must have been proto-Indo-European if both are roughly correlated to caste in modern India.

    The authors would have been better off using a components derived from ancient samples (Yamnaya, Sintashta, Iran_ChL, Iran_Hotu etc.) with moderns where we currently lack the ancestral sources (f.ex. Paniya for ASI).

    That there was a South Asian presence in Xinjiang around 4000 YBP in the Tarim was already established by the discovery of mtdna M5, which is indisputably of South Asian origin, among the ancient samples from Xiaohe (dated to between 4000 - 3500 YBP).

    So, the steppe origin theory of Tocharians looks shaky.
    Please go through the original study once again:

    Quote Originally Posted by Li et al. 2015
    In addition, we found one individual with the Indian lineage M5 [48]. Nowadays, the M5 variant observed in this study is found mainly in south and southwest Asia. The presence of hgs U7 and M5 in the Xiaohe people suggests that populations of west/south Asia contributed to the gene pool of the Tarim Basin during the Bronze Age.
    ^ A single individual, and the authors are stating this variant is also found in West Asia. I don't know much about M5, but assuming this particular lineage must have come directly from the Subcontinent (as you hold) isn't a neutral inference. The onus is on you to prove that particular variant of M5 is South Asian and not West Asian.

    Asserting I'm not reading your post correctly, then completely omitting the above regarding that one Xiaohe M5. Selective reading and hypocrisy makes you a naughty one!

  9. The Following 4 Users Say Thank You to DMXX For This Useful Post:

     Awale (06-18-2017), Coldmountains (06-18-2017), Michał (06-18-2017), Tsakhur (06-18-2017)

  10. #16
    Registered Users
    Posts
    4,672

    Quote Originally Posted by DMXX View Post
    ...


    ^ A single individual, and the authors are stating this variant is also found in West Asia. I don't know much about M5, but assuming this particular lineage must have come directly from the Subcontinent (as you hold) isn't a neutral inference. The onus is on you to prove that particular variant of M5 is South Asian and not West Asian.

    Asserting I'm not reading your post correctly, then completely omitting the above regarding that one Xiaohe M5. Selective reading and hypocrisy makes you a naughty one!
    On the flip side it may be more than a single individual as other lines such as U7 and R* could also be from South Asia.

    "Mitochondrial DNA analysis showed that the Xiaohe people carried both the East Eurasian haplogroup (C) and the West Eurasian haplogroups (H and K), whereas Y chromosomal DNA analysis revealed only the West Eurasian haplogroup R1a1a in the male individuals."

    Notably most R1a1 samples carried the East Eurasian haplogroup C4 and others were R* ("The results showed that they are related neither to the West Eurasian haplogroups UK, TJ, HV, R11 and R1, nor to the East Eurasian haplogroups B and F. So we designated them as haplogroup R* temporarily.")

    111
    223-298-309-327 C4
    R1a1a

    115
    298-327 C4
    R1a1a

    120
    189-192-311 R*
    R1a1a

    121
    183-189-192-311 R*
    R1a1a

    136
    298-327 C4
    R1a1a

    139
    298-327 C4
    R1a1a

  11. The Following 2 Users Say Thank You to parasar For This Useful Post:

     Baltimore1937 (06-20-2017), Coldmountains (06-19-2017)

Page 2 of 2 FirstFirst 12

Similar Threads

  1. ANE as an admixture
    By ren in forum Ancient (aDNA)
    Replies: 147
    Last Post: 07-02-2017, 01:35 PM
  2. Replies: 4
    Last Post: 05-13-2017, 08:06 PM
  3. Replies: 4
    Last Post: 10-27-2015, 09:47 PM
  4. Replies: 38
    Last Post: 04-11-2014, 04:47 PM
  5. Ancient Admixture in Human History (Patterson et al 2012)
    By Jean M in forum Autosomal (auDNA)
    Replies: 1
    Last Post: 09-07-2012, 10:45 PM

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •