Page 1 of 2 12 LastLast
Results 1 to 10 of 18

Thread: Estimating amount of East and West Eurasian ancestries of Amerindians using Global 25

  1. #1
    Registered Users
    Posts
    513
    Sex
    Omitted

    Estimating amount of East and West Eurasian ancestries of Amerindians using Global 25

    Not sure if this is the appropriate section for this..

    So I tried to the model Amerindians as a mix between East Eurasian (Devil's Gate is used here as a proxy) and West Eurasian (Corded_Ware_DEU) through Global 25 nMonte and here are the results. Formal stats would give more precise and accurate answers but I don't know how to utilize them so this is the most I can do for now.

    There is one peculiar observation though as the East and West Eurasian they are scoring in this run does not seem to correlate to their 30-40% ANE as Pima and Huichol should score a lot less West Eurasian than Karitiana as the former score less ANE than the latter. But almost every group except Chipewyan and Amerindian_North are literally getting the same amounts of West Eurasian (28-29%) Are there factors behind this?

    "sample": "Chipewyan:Average", (Dene/Athabaskan tribe from Canada; they have lower West Eurasian than other Amerindians which is likely due to recent Siberian gene flow)
    "fit": 31.8138,
    "RUS_Devils_Gate_Cave_N": 74.17,
    "Corded_Ware_DEU": 25.83,


    "sample": "Amerindian_North:Average", (the samples are from the USA but don't know which tribe/group)
    "fit": 34.516,
    "RUS_Devils_Gate_Cave_N": 74.17,
    "Corded_Ware_DEU": 25.83
    ,



    "sample": "Pima:Average",
    "fit": 45.4868,
    "RUS_Devils_Gate_Cave_N": 71.67,
    "Corded_Ware_DEU": 28.33,

    "sample": "Huichol:Average",
    "fit": 45.7884,
    "RUS_Devils_Gate_Cave_N": 70.83,
    "Corded_Ware_DEU": 29.17,

    "sample": "Mixtec:Average", (The Mixtec seem to have around 4-5% Euro admix when using the ancient Belize genome and Basque Spanish as proxies)
    "fit": 44.8149,
    "RUS_Devils_Gate_Cave_N": 69.17,
    "Corded_Ware_DEU": 30.83,

    "sample": "Karitiana:Average",
    "fit": 53.4244,
    "RUS_Devils_Gate_Cave_N": 71.67,
    "Corded_Ware_DEU": 28.33,

    "sample": "Surui:Average",
    "fit": 54.3301,
    "RUS_Devils_Gate_Cave_N": 70.83,
    "Corded_Ware_DEU": 29.17,

    "sample": "Aymara:Average",
    "fit": 49.9763,
    "RUS_Devils_Gate_Cave_N": 70.83,
    "Corded_Ware_DEU": 29.17,

    "sample": "Colla:Average", (they seem to have around 4-5% Euro admix when tried to run them with ARG_Aconcagua and Basque Spanish)
    "fit": 48.4859,
    "RUS_Devils_Gate_Cave_N": 68.33,
    "Corded_Ware_DEU": 31.67,

  2. #2
    Registered Users
    Posts
    513
    Sex
    Omitted

    It seems that if we can eliminate the genetic drift, Native Americans would literally plot almost in the same position with some Turkic Central Asians and Turkic Siberians such as Altaian, Khakass_Kachin or even Kirghiz on a 3rd dimension East-West Eurasian and African Global population PCA thanks to their similar proportions of East and West Eurasian ancestries.

    The second trial, I attempt to utilize a different West Eurasian proxy (Bell_Beaker_Bavaria, the last one I used was also from Germany). For some reason, it decrease the West Eurasian amount of Natives.

    "sample": "Mongolian:Average",
    "fit": 4.7389,
    "RUS_Devils_Gate_Cave_N": 80,
    "Bell_Beaker_Bavaria": 20,


    "sample": "Pima:Average",
    "fit": 46.0149,
    "RUS_Devils_Gate_Cave_N": 74.17,
    "Bell_Beaker_Bavaria": 25.83,

    "sample": "Altaian:Average", (Turkic ethnic group from South Siberia; it is interesting because on PCA they are actually located 3% more western-shifted from where the Amerindians are plotted)
    "fit": 6.6794,
    "RUS_Devils_Gate_Cave_N": 73.33,
    "Bell_Beaker_Bavaria": 26.67,


    "sample": "Karitiana:Average",
    "fit": 53.8515,
    "RUS_Devils_Gate_Cave_N": 74.17,
    "Bell_Beaker_Bavaria": 25.83,

    "sample": "Khakass_Kachins:Average", (Turkic tribe from South Siberia)
    "fit": 7.4516,
    "RUS_Devils_Gate_Cave_N": 73.33,
    "Bell_Beaker_Bavaria": 26.67,

    "sample": "Chipewyan:Average", (They seem to have recent Siberian ancestry)
    "fit": 32.4336,
    "RUS_Devils_Gate_Cave_N": 76.67,
    "Bell_Beaker_Bavaria": 23.33,

    "sample": "Colla:Average", (They are from Argentina and related to Aymaras; seem to have around 5% Euro admix when using )
    "fit": 48.9093,
    "RUS_Devils_Gate_Cave_N": 70.83,
    "Bell_Beaker_Bavaria": 29.17,

    "sample": "Kirghiz:Average", (looks like an Amerindian with 6-7% Euro admix would cluster close to them)
    "fit": 4.5868,
    "RUS_Devils_Gate_Cave_N": 68.33,
    "Bell_Beaker_Bavaria": 31.67,


    "sample": "Nenets:Average", (an Uralic tribe from Arctic part of Russia, they have a lot of ANE ancestry)
    "fit": 19.0829,
    "RUS_Devils_Gate_Cave_N": 68.33,
    "Bell_Beaker_Bavaria": 31.67,


    Not sure whether using Cord Ware or Bell Beaker is better but it seems the former has more ENA admix than the former (trying to run them with Sardinian and Ulchi)

    I know formal stats is overall a lot more accurate but this also seems mindblowing to me lol.

  3. #3
    Registered Users
    Posts
    513
    Sex
    Omitted

    Opinions please? How much East and West Eurasian do they really have? Is what I am doing accurate and the right method to figure out the answer? Also would they really cluster with/plot close to some Turkic Central Asians, Turkic Siberians and some Yeniseians, Samoyeds like Altaians, Khakass, Kyrgyz, Ket, Selkup, Nenets, etc. on Global East-West-Africa PCA if the genetic drift can be eliminated?

  4. #4
    Registered Users
    Posts
    1,112
    Sex
    Location
    Central Florida
    Ethnicity
    Aegean Greek + NW Euro
    Nationality
    American
    Y-DNA (P)
    J-L26
    mtDNA (M)
    J1b1a

    Greece United States of America
    Any particular reason you're using West Eurasian groups instead of Malta boy or AG3? West Eurasians are not going to be good proxies for the non-ENA ancestry in Amerinds because of their Basal Eurasian ancestry, something ancestral Amerinds would not have. Of course, many modern Amerinds often have European ancestry which confounds things a bit. I would try to focus on the pre-Columbian American samples we have to eliminate that problem.

    We don't have a very good proxy for Amerinds' East Asian ancestry either. Devils Gate is too young and Tianyuan too old. Kolyma is closest to what the Amerinds actually have but it's not perfect either.
    Last edited by Michalis Moriopoulos; 02-07-2020 at 05:52 PM.
    Ελευθερία ή θάνατος.

  5. The Following 6 Users Say Thank You to Michalis Moriopoulos For This Useful Post:

     Helgenes50 (02-08-2020),  JoeyP37 (02-07-2020),  pgbk87 (03-27-2020),  Sapporo (07-11-2020),  Sorcelow (02-08-2020),  Tsakhur (02-07-2020)

  6. #5
    Registered Users
    Posts
    513
    Sex
    Omitted

    Quote Originally Posted by Michalis Moriopoulos View Post
    Any particular reason you're using West Eurasian groups instead of Malta boy or AG3? West Eurasians are not going to be good proxies for the non-ENA ancestry in Amerinds because of their Basal Eurasian ancestry, something ancestral Amerinds would not have. Of course, many modern Amerinds often have European ancestry which confounds things a bit. I would try to focus on the pre-Columbian American samples we have to eliminate that problem.

    We don't have a very good proxy for Amerinds' East Asian ancestry either. Devils Gate is too young and Tianyuan too old. Kolyma is closest to what the Amerinds actually have but it's not perfect either.
    (

    The reason why I am using West Eurasian pops as proxies is because that I have learnt from many people that Malta/AG3 descends from Yana who are actually around 3/4 West Eurasian (Gravettian-derived) with around 1/4 East Asian in genetic composition. Moreover, I heard Native Americans are around 30-40% ANE/AG3 so thus they would be approximately 23-30% West Eurasian. Therefore, if ANE is really primarily West Eurasian, I thought utilizing direct West Eurasian pops would be better than Malta or AG3. Even though, now I know that West Eurasian groups also possess Basal Eurasian ancestry.

    Please correct and educate me if my understanding regarding this topic is wrong. I am pretty much a layperson and amateur in the subjects of genetics.

    Interesting. Does this mean that ANE also lack Basal Eurasian?

    I see. Will attempt to run the models again using what you have suggested.
    Last edited by Tsakhur; 02-07-2020 at 06:41 PM.

  7. #6
    Registered Users
    Posts
    513
    Sex
    Omitted

    More opinions please?

  8. #7
    Registered Users
    Posts
    1,112
    Sex
    Location
    Central Florida
    Ethnicity
    Aegean Greek + NW Euro
    Nationality
    American
    Y-DNA (P)
    J-L26
    mtDNA (M)
    J1b1a

    Greece United States of America
    Quote Originally Posted by Tsakhur View Post
    The reason why I am using West Eurasian pops as proxies is because that I have learnt from many people that Malta/AG3 descends from Yana who are actually around 3/4 West Eurasian (Gravettian-derived) with around 1/4 East Asian in genetic composition.
    I'm not sure if they descend directly from Yana as opposed to coming from a common North Eurasian root. In any case, there is a difference between modern West Eurasians and ancestral Western Crown Eurasian groups like Villabruna, Sunghir, and the non-ENA part of ANE. That difference is "deep" ancestry in modern West Eurasians that doesn't exist in those super ancient populations (or if it does exist in them, it's very low). We call this deep ancestry Basal Eurasian. There are clever ways to make the statistical need for Basal Eurasian disappear, but the orthodox assumption for now is that such a population did in fact exist. So if we define "West Eurasian" as Western Crown Eurasian + Basal Eurasian, the oldest population we know of that fits that criterion is Dzudzuana (26,000 years old).

    Quote Originally Posted by Tsakhur View Post
    Moreover, I heard Native Americans are around 30-40% ANE/AG3
    Yes, that's true.

    Quote Originally Posted by Tsakhur View Post
    Interesting. Does this mean that ANE also lack Basal Eurasian?
    Yes. Either that or it's at a very low frequency.

    Again, this isn't set in stone. Our models could change dramatically with new information. But this is where we are right now.
    Ελευθερία ή θάνατος.

  9. The Following User Says Thank You to Michalis Moriopoulos For This Useful Post:

     Tsakhur (02-08-2020)

  10. #8
    Registered Users
    Posts
    1,742
    Sex
    Omitted

    Global25 or any PCA really needs proximate sources to be accurate. The 'west Eurasian' is being pushed down in percentage because Corded_Ware shares drift with other components in the PCA that the true West Eurasian source (MA1/AG3) does not have.
    Collection of 14,000 d-stats: Hidden Content Part 2: Hidden Content Part 3: Hidden Content PM me for d-stats, qpadm, qpgraph, or f3-outgroup nmonte models.

  11. The Following User Says Thank You to Kale For This Useful Post:

     Tsakhur (02-08-2020)

  12. #9
    Registered Users
    Posts
    4,612
    Sex

    Always model Amerindians as Siberian Neolithics + MA1/AG3 + English + Iberian. That should give you sensible results using the Global25. But you need to pick the right ancient Siberians. I'm not sure which ones work best.

    Corded Ware is a very strange reference pop to use in this context.

  13. The Following User Says Thank You to Generalissimo For This Useful Post:

     Tsakhur (02-08-2020)

  14. #10
    Registered Users
    Posts
    513
    Sex
    Omitted

    Quote Originally Posted by Generalissimo View Post
    Always model Amerindians as Siberian Neolithics + MA1/AG3 + English + Iberian. That should give you sensible results using the Global25. But you need to pick the right ancient Siberians. I'm not sure which ones work best.

    Corded Ware is a very strange reference pop to use in this context.
    Hmm thank you for the feedback regarding this.

    So we can gauge the amount of West Eurasian ancestry in Amerindians through the combination of MA1/AG3+English+Iberian? But would not English and Iberian reflect the modern Euro admix in Native Americans instead of ancient West Eurasian ancestry that they acquired from their ANE forefathers?

    Can you give some examples of ancient Siberian proxies that I can attempt for my modeling?

Page 1 of 2 12 LastLast

Similar Threads

  1. Replies: 5
    Last Post: 12-11-2019, 09:43 AM
  2. Replies: 10
    Last Post: 01-03-2019, 11:43 PM
  3. How much ancient West Eurasian ancestry do Amerindians have?
    By Tsakhur in forum Autosomal (auDNA)
    Replies: 4
    Last Post: 09-18-2017, 12:06 PM
  4. Replies: 4
    Last Post: 10-10-2015, 02:35 AM
  5. Replies: 52
    Last Post: 02-18-2015, 10:14 AM

Posting Permissions

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