Page 2 of 14 FirstFirst 123412 ... LastLast
Results 11 to 20 of 139

Thread: Is Basal Eurasian real or an fstats artifact?

  1. #11
    Registered Users
    Posts
    324
    Sex
    Omitted

    Quote Originally Posted by Mnemonics View Post
    The idea of a Basal Eurasian population which split off main OOA population before they admixed with Neanderthals has been a pretty popular explanation for the depleted Neanderthal alleles present in various ancient Near Eastern populations (Natufians, Iran_N etc).
    Have you ever tried emailing Nick Patterson or other professionals in the field (Iosif Lazaridis, Martin Petr, Pontus Skoglund, etc) and seeing if they have any thoughts on this? I've always thought the f-stats issue with Africans was really weird (in addition to this Basal thing there's also the fact that I think f-stats make Africans look way more unrelated to each other then they really are) and I've never seen a good explanation here or anywhere else as to why.

    Also, a while back I remember you seemed to have had some success modeling Africans as being Ust-Ishim-like, implying some real ancient Eurasian back-migration to Africa. I'm wondering if you're still open to that idea (have you tried using the Bacho Kiro or Zlaty Kun samples to model Africans, has that produced anything interesting?) and if so, what implication would that have on the BE phenomenon?
    Last edited by TuaMan; 09-12-2021 at 11:53 PM.

  2. The Following 3 Users Say Thank You to TuaMan For This Useful Post:

     Gadzooks (09-15-2021),  Megalophias (09-13-2021),  Mnemonics (09-13-2021)

  3. #12
    Registered Users
    Posts
    247
    Sex
    Location
    Canada
    Ethnicity
    Somali
    Y-DNA (P)
    T-L208
    mtDNA (M)
    N1b2

    Canada Somaliland
    Quote Originally Posted by Kale View Post
    Hmmm I think you may have hit on something unexpected here... For every African population I've seen Chimp African1 Eurasian African2 is always negative (even when African2 is Mota) because of multiple deep layers of ancestry in Africa presumably. Why doesn't PastoralN show this?

    The only one I've found that's the opposite is Dinka.
    1 Chimp~ Mbuti.DG Mongolia_N_East Dinka.DG 4.32955
    2 Chimp~ Yoruba.DG Mongol~ Dink~ 3.71215
    3 Chimp~ South_Africa_2000BP.SG Mongol~ Dink~ 0.221423
    I assumed it was a result of the Eurasian bottleneck causing there to be much more shared alleles between Eurasian + SSA populations and their Eurasian sources relative to other Africans. It seems to get a bit better when you use more recently related individuals


    Code:
    result:      Chimp Tanzania_PastoralN   BedouinB  Mnemonics     -0.008973      0.000414   -21.673   40446  47226 755515 
    result:      Chimp  Mnemonics   BedouinB Tanzania_PastoralN     -0.009162      0.000425   -21.542   40446  47368 755515 
    result:      Chimp Kenya_PastoralN   BedouinB  Mnemonics     -0.008652      0.000383   -22.582   47661  55341 887627 
    result:      Chimp  Mnemonics   BedouinB Kenya_PastoralN     -0.009372      0.000416   -22.557   47661  55981 887627 
    
    result:      Chimp DrobbahImp   BedouinB  Mnemonics     -0.005708      0.000523   -10.909   52681  58106 950504 
    result:      Chimp  Mnemonics   BedouinB DrobbahImp     -0.006991      0.000511   -13.681   52681  59326 950504
    It still remains significantly negative.

    Quote Originally Posted by Nebuchadnezzar II View Post
    What happens when you use Mota? I.e (Chimp, Mota, Eurasian, Natufian)
    Code:
    result:      Chimp       Mota        WHG   Natufian     -0.001137      0.000449    -2.533   23067  23643 506210 
    result:      Chimp       Mota        EHG   Natufian     -0.000480      0.000460    -1.043   23360  23600 501478 
    result:      Chimp       Mota        CHG   Natufian     -0.000750      0.000489    -1.534   23485  23864 506231 
    result:      Chimp       Mota       PPNB   Natufian     -0.000463      0.000411    -1.127   19477  19688 456584 
    result:      Chimp       Mota Anatolia_N   Natufian     -0.000765      0.000359    -2.131   22364  22751 506039

  4. #13
    Registered Users
    Posts
    499
    Sex
    Location
    Rome
    Ethnicity
    Veneto\Toscano
    Nationality
    Italian
    Y-DNA (P)
    R1b-U152-L2-Z367
    mtDNA (M)
    T1a1

    Republic of Venice Italy Veneto Italy Tuscany Italy 1861-1946 European Union Italy
    SSA pops are far more unrelated to each other than Eurasians. That's why Africans are not a great match for each other with f-stats. Using Yoruba for Natufians thinking that is a relevant or representative sample of an hypothetical ANA-like population that maybe lived 50k years ago, is simply bad methodology. That population might have been equally related to OOA compared to Yoruba. Mota is a great example of that. Convince me otherwise.

  5. The Following 5 Users Say Thank You to Ariel90 For This Useful Post:

     beyoku (09-13-2021),  Mansamusa (09-13-2021),  pgbk87 (Yesterday),  taharqa (09-14-2021),  whynot (09-17-2021)

  6. #14
    Registered Users
    Posts
    247
    Sex
    Location
    Canada
    Ethnicity
    Somali
    Y-DNA (P)
    T-L208
    mtDNA (M)
    N1b2

    Canada Somaliland
    Quote Originally Posted by TuaMan View Post
    Have you ever tried emailing Nick Patterson or other professionals in the field (Iosif Lazaridis, Martin Petr, Pontus Skoglund, etc) and seeing if they have any thoughts on this? I've always thought the f-stats issue with Africans was really weird (in addition to this Basal thing there's also the fact that I think f-stats make Africans look way more unrelated to each other then they really are) and I've never seen a good explanation here or anywhere else as to why.

    Also, a while back I remember you seemed to have had some success modeling Africans as being Ust-Ishim-like, implying some real ancient Eurasian back-migration to Africa. I'm wondering if you're still open to that idea (have you tried using the Bacho Kiro or Zlaty Kun samples to model Africans, has that produced anything interesting?) and if so, what implication would that have on the BE phenomenon?
    Mota and the Southeast African hunter-gather cline seem to have some real Eurasian ancestry. They also have some very deep ancestry which pulls them to South Africans.


    Quote Originally Posted by Ariel90 View Post
    SSA pops are far more unrelated to each other than Eurasians. That's why Africans are not a great match for each other with f-stats. Using Yoruba for Natufians thinking that is a relevant or representative sample of an hypothetical ANA-like population that maybe lived 50k years ago, is simply bad methodology. That population might have been equally related to OOA compared to Yoruba. Mota is a great example of that. Convince me otherwise.
    If you cannot get a great match for Africans in fstats then how can you rule out African admixture in the Eurasian admixed populations?

    If you can't rule out African admixture then what is the most parsimonious explanation for the lowered Neanderthal alleles and pull towards African populations in "Basal rich " populations?
    Last edited by Mnemonics; 09-13-2021 at 02:18 AM.

  7. The Following 2 Users Say Thank You to Mnemonics For This Useful Post:

     Mansamusa (Yesterday),  taharqa (09-14-2021)

  8. #15
    Registered Users
    Posts
    324
    Sex
    Omitted

    Quote Originally Posted by Ariel90 View Post
    SSA pops are far more unrelated to each other than Eurasians. That's why Africans are not a great match for each other with f-stats. Using Yoruba for Natufians thinking that is a relevant or representative sample of an hypothetical ANA-like population that maybe lived 50k years ago, is simply bad methodology. That population might have been equally related to OOA compared to Yoruba. Mota is a great example of that. Convince me otherwise.
    I know this seems to be conventional wisdom with a lot of people but I don't buy it. A while back I was communicating with Kale over PM and he shared some D-Stats (I unfortunately don't have access to them at the moment) where Yoruba-Igbo (I believe the stat was in the form of Chimp Yoruba Eurasian Igbo or something like that) had barely significant Z-score affinities of 2 or 3 or something, whereas the stat Chimp French African German had Zs of like 20 or 30. There's simply no way Yorubas and Igbos are so diverged that they're just barely more related to each other than they are to random Europeans.
    Last edited by TuaMan; 09-13-2021 at 03:04 AM.

  9. The Following 2 Users Say Thank You to TuaMan For This Useful Post:

     Megalophias (09-13-2021),  Mnemonics (09-13-2021)

  10. #16
    Registered Users
    Posts
    247
    Sex
    Location
    Canada
    Ethnicity
    Somali
    Y-DNA (P)
    T-L208
    mtDNA (M)
    N1b2

    Canada Somaliland
    Quote Originally Posted by TuaMan View Post
    I know this seems to be conventional wisdom with a lot of people but I don't buy it. A while back I was communicating with Kale over PM and he shared some D-Stats (I unfortunately don't have access to them at the moment) where Yoruba-Igbo (I believe the stat was in the form of Chimp Yoruba Eurasian Igbo or something like that) had barely significant Z-score affinities of 2 or 3 or something, whereas the stat Chimp French African German had Zs of like 20 or 30 or something. There's simply no way Yorubas and Igbos are so diverged that they're just barely more related to each other than they are to random Europeans.
    What ever the issue with direct fstats is f4Ratios tend to be fine at distinguishing the relative relationships between African populations which indicates that what ever is causing the skewed stats is relatively uniform.

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

     pegasus (09-13-2021),  TuaMan (09-13-2021)

  12. #17
    Registered Users
    Posts
    72
    Sex

    What do you think of the 2018 Cabrera paper, "Carriers of mitochondrial DNA macrohaplogroup L3 basal lineages migrated back to Africa from Asia around 70,000 years ago"?

    Humans left Africa over 100k years ago, bringing pre-L3 and undifferentiated CT with them. CT, prior to mixing with the Neanderthals, split into CF and DE somewhere in Asia, with the DE's staying close to Southwest Asia without mixing with the Neanderthals. DE's split into D's and E's with E's and a rare D migating back to Africa around 30-40 kya. Some D's migrated east, picking up Neanderthal and Denisovan ancestry on the way, whereas the E's, still with no Neanderthal ancestry, stayed in/migrated to Southwest Asia, later contributing to the Natufians and to the Basal Eurasian component of the Dzudzuana-type people to the north of the Fertile Crescent. The Natufians got their Neanderthal ancestry from the mating networks with the Dzudzuana people to the north of them and from the Aurignacians.

    Makes sense or unrealistic?
    Last edited by Woozler; 09-13-2021 at 05:38 AM.

  13. The Following 3 Users Say Thank You to Woozler For This Useful Post:

     aganmu (09-13-2021),  etrusco (09-13-2021),  thejkhan (09-13-2021)

  14. #18
    Registered Users
    Posts
    146
    Sex
    Ethnicity
    Finnish

    Quote Originally Posted by TuaMan View Post
    I think f-stats make Africans look way more unrelated to each other then they really are
    I thought that the f2 distances between African populations were surprisingly small relative to Eurasian populations. Maybe I did something wrong, but based on f2 distances that I calculated between populations in 1240K+HO, even the distance of Yoruba to Herero is around the same as the distance of French to Russian_Arkhangelsk_Leshukonsky. And the distance of Yoruba to Khomani is around the same as the distance of French to Ket.


  15. The Following 5 Users Say Thank You to Nganasankhan For This Useful Post:

     Archetype0ne (09-13-2021),  Mnemonics (09-13-2021),  Tsakhur (09-14-2021),  TuaMan (09-13-2021),  whynot (09-17-2021)

  16. #19
    Registered Users
    Posts
    499
    Sex
    Location
    Rome
    Ethnicity
    Veneto\Toscano
    Nationality
    Italian
    Y-DNA (P)
    R1b-U152-L2-Z367
    mtDNA (M)
    T1a1

    Republic of Venice Italy Veneto Italy Tuscany Italy 1861-1946 European Union Italy
    Quote Originally Posted by TuaMan View Post
    I know this seems to be conventional wisdom with a lot of people but I don't buy it. A while back I was communicating with Kale over PM and he shared some D-Stats (I unfortunately don't have access to them at the moment) where Yoruba-Igbo (I believe the stat was in the form of Chimp Yoruba Eurasian Igbo or something like that) had barely significant Z-score affinities of 2 or 3 or something, whereas the stat Chimp French African German had Zs of like 20 or 30. There's simply no way Yorubas and Igbos are so diverged that they're just barely more related to each other than they are to random Europeans.
    This is starting to become a conspiracy theory, Africans are hardly genetically related based on the tools that we are using, these tools have been developed by leading scientits, ultimately is DNA reads and math, right? Where is the potential flaw? Basal per se falls well outside modern African variations, but it doesn't mean that Natufians had no African admix in a geographical sense, because they almost certainly did, but the notion of a distinct SSA general lineage is not supported by anything, it's pseudo-science. And when it comes to East-Africa, well, it's even more obvious...

  17. The Following 2 Users Say Thank You to Ariel90 For This Useful Post:

     beyoku (09-13-2021),  Riverman (09-13-2021)

  18. #20
    Registered Users
    Posts
    2,672
    Sex

    Quote Originally Posted by Woozler View Post
    What do you think of the 2018 Cabrera paper, "Carriers of mitochondrial DNA macrohaplogroup L3 basal lineages migrated back to Africa from Asia around 70,000 years ago"?

    Humans left Africa over 100k years ago, bringing pre-L3 and undifferentiated CT with them. CT, prior to mixing with the Neanderthals, split into CF and DE somewhere in Asia, with the DE's staying close to Southwest Asia without mixing with the Neanderthals. DE's split into D's and E's with E's and a rare D migating back to Africa around 30-40 kya. Some D's migrated east, picking up Neanderthal and Denisovan ancestry on the way, whereas the E's, still with no Neanderthal ancestry, stayed in/migrated to Southwest Asia, later contributing to the Natufians and to the Basal Eurasian component of the Dzudzuana-type people to the north of the Fertile Crescent. The Natufians got their Neanderthal ancestry from the mating networks with the Dzudzuana people to the north of them and from the Aurignacians.

    Makes sense or unrealistic?
    I think the splits go like this:
    1. Archaic <-> Basal African
    2. Basal African (yDNA A, B primarily) <-> Neo-African (rest)
    3. Ancient North African <-> Eurasian
    4. Basal Eurasian <-> Crown Eurasian

    1 + 2 happened in Africa almost for sure, but 3 + 4 could have happened in North East Africa-Sinai-Levante-Arabia-Mesopotamia, with possible forth and back migrations. The Shum Laka paper is the only really relevant on the matter for Africa itself and it shows that haplogroup E was not there yet very recently, but spread with ANA to Subsaharan Africa. YDNA E and mtDNA L3 spread with ANA into Subsaharan Africa fairly recently.
    The other question is, which paternal lineages represent more recent Basal Eurasian, I'd say it could be E1b1b primarily. But we need more ancient DNA from the Near East, pre-Natufian and from Arabia, especially Yemen and of course Egypt and Sinai too.

    Iberomaurusians are too late and just a mixed population of West-Basal Eurasian with ANA. How and where that came up is open to debate. Like suggested above, I think multiple forth and back migrations between Egypt-Sinai-Levante-Arabia are possible and would further complicate the picture. However, much of Africa, practically all of Subsaharan Africa, got ANA and E/L3 ancestry just fairly recently, that's for sure.

  19. The Following 5 Users Say Thank You to Riverman For This Useful Post:

     Ahmed Achik (09-18-2021),  Ariel90 (09-13-2021),  etrusco (09-13-2021),  pgbk87 (Yesterday),  TuaMan (09-13-2021)

Page 2 of 14 FirstFirst 123412 ... LastLast

Similar Threads

  1. What is Basal Eurasian?
    By Tsakhur in forum Autosomal (auDNA)
    Replies: 204
    Last Post: 07-19-2021, 12:30 AM
  2. Replies: 3
    Last Post: 06-15-2020, 02:32 PM
  3. Is haplogroup M1 Basal Eurasian?
    By NetNomad in forum M
    Replies: 25
    Last Post: 10-12-2018, 06:27 PM
  4. Is there a need for Basal Eurasian now?
    By ren in forum Ancient (aDNA)
    Replies: 0
    Last Post: 10-14-2017, 02:58 AM
  5. Basal Eurasian and ASI Split
    By Dr_McNinja in forum Ancient (aDNA)
    Replies: 158
    Last Post: 11-06-2015, 07:18 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
  •