Page 3 of 7 FirstFirst 12345 ... LastLast
Results 21 to 30 of 63

Thread: Why are most east Euros further from MLBA steppe than northern/western/central Euros?

  1. #21
    Registered Users
    Posts
    317
    Sex

    Extra genetic drift probably matters. Though like I say, effect shows up even in stats that supposedly insensitive to drift like f3 outgroup, so that can't be all of it.

    I don't see much evidence of a bottleneck and so extra genetic drift in a Common Era Slavic expansion though; that just seems like maybe a population expansion from a relatively low starting size with no bottlenecks around that time (hence no real extra genetic drift). I think any bottlenecks involved probably happened in the genesis of Baltic_BA (or whatever more widely distributed population Baltic_BA is actually a subset of). Your mileage may vary.

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

     DMXX (01-09-2019),  Hando (01-26-2019),  Michał (01-09-2019),  Ryukendo (01-19-2019)

  3. #22
    Bronze Class Member
    Posts
    1,616
    Sex

    Turkmenistan Suriname South Africa Costa Rica Norway Australia
    Quote Originally Posted by Helves View Post
    I agree regarding the CWC, but my point is that it's not the various Steppe or Steppe-rich populations that are a bad fit for modern day NE Euros but their non-IE part that has probably not been sampled yet. Because even using the Sredny Stog R1a sample or the other CWC available the fit isn't nearly as good as for NW Euros.
    This is what I was initially suspecting-that there might be some unknown population that bred with Yamnaya/Corded Ware to create Baltic-BA which then gave rise to Balts/east Slavs and that relic population is not shared with Sintashta, which results in worse fits for them compared to NW/northern Euros.

    Drift is a possibility but I have seen on G25, how isolating the proper ancestral components, now matter how long the "test group" in question has been isolated(I was able to get a low fit for Andaman islanders even), can provide a decent fit. I dont see why this would not happen with Balto-Slavs.

    I don't buy the idea that there is something going on with the ratios of WHG and EHG because that can easily be tested out when you add those in(along with other HGs). Any extra WHG or EHG or whatever else not captured by the main groups(Globular amphora+Yamnaya) should show up.

    Code:
    1	Baltic_HG +Barcin_N +EHG +Globular_Amphora +Latvia_HG +Latvia_MN +Narva_Lithuania +SHG +WHG +Yamnaya_Samara	German:Average		2.2526	Open Map	0	0.83	0.83	55	0	0	0	0	0	43.33
    2	Baltic_HG +Barcin_N +EHG +Globular_Amphora +Latvia_HG +Latvia_MN +Narva_Lithuania +SHG +WHG +Yamnaya_Samara	Lithuanian:Average		6.7004	Open Map	0	12.5	0	21.67	5.83	5	13.33	1.67	0	40
    I know this is a lot of groups that might overlap but Im trying to get an idea of where the ancestry comes from. And look at that, the Lithuanians' fit is still much worse than the Germans in spite of all the groups added, but with Baltic_BA that changes:

    Code:
    1	Baltic_BA +Baltic_HG +Barcin_N +EHG +Globular_Amphora +Latvia_HG +Latvia_MN +Narva_Lithuania +SHG +WHG +Yamnaya_Samara	German:Average		1.8161	Open Map	15.83	0	7.5	0	40.83	0	0	0	0	0	35.83
    2	Baltic_BA +Baltic_HG +Barcin_N +EHG +Globular_Amphora +Latvia_HG +Latvia_MN +Narva_Lithuania +SHG +WHG +Yamnaya_Samara	Lithuanian:Average		1.6667	Open Map	75	0	12.5	0	3.33	0	0	0	0	0	9.17
    For the record, Lithuanians and Germans are equally close to Yamnaya:
    1 Yamnaya_Samara German:Average 14.6251 Open Map 100
    2 Yamnaya_Samara Lithuanian:Average 14.7668 Open Map 100

    But not to Sintashta:
    1 Sintashta_MLBA German:Average 6.0995 Open Map 100
    2 Sintashta_MLBA Lithuanian:Average 7.9403 Open Map 100

    So I'm inclined towards believing that there is indeed some sort of unaccounted for non-IE population that Balts and east Slavs mixed with, which is not found in Sintashta or in northern/northwestern Europeans. If that's the case, then Sintashta ironically would have a more similar ethnogenesis to northern/northwestern Europeans in spite of being located at the far southeastern edge of Europe.

    Keep in mind my opinion should be taken with a grain of salt as Im no expert in European archaeogenetics. This is just what it looks like to me at this moment, until I see further evidence.
    Last edited by Censored; 01-08-2019 at 10:07 PM.

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

     Hando (01-26-2019),  Helves (01-08-2019),  palamede (01-09-2019),  Rafe (01-11-2019),  Sorcelow (01-18-2019)

  5. #23
    Indefinite Hiatus
    Posts
    3,145
    Sex
    Y-DNA
    R2a*-M124 (L295-)
    mtDNA
    D4j5*

    England
    You're right (Eterne), though I had founder effects in mind rather than bottlenecks. The regionalisation of Y-DNA R1a-Z283 subclades in Europe (I have M458 and Z280 in mind) is potential circumstantial evidence in favour of this.

    Quote Originally Posted by Helves View Post
    Yes, but will the genetic drift have this big of a role that I can get a fit under 3(with nMonte) for NW Euros using the various Steppe-rich samples(BB, CWC) available whilst the same runs give terrible fits for Balts and other NE Euros?
    How can modern day Balto-Slavs rich in R1a-M417 not share the same amount of drift with the CWC samples the same way NW Euros do with the Bell Beakers?
    Probably not the main reason. I only stated that it's one factor. We don't have the appropriate ancient samples to model Slavic populations yet (particularly for East Slavs - Closest thing so far is CWC Baltic).

  6. The Following 7 Users Say Thank You to DMXX For This Useful Post:

     Agamemnon (01-08-2019),  Eterne (01-08-2019),  Hando (01-26-2019),  Helves (01-08-2019),  Michał (01-09-2019),  Ruderico (01-09-2019),  Táltos (01-08-2019)

  7. #24
    Registered Users
    Posts
    2,319
    Location
    Gonur Tepe

    Afghanistan Jammu and Kashmir United States of America Canada
    Quote Originally Posted by DMXX View Post
    You're right (Eterne), though I had founder effects in mind rather than bottlenecks. The regionalisation of Y-DNA R1a-Z283 subclades in Europe (I have M458 and Z280 in mind) is potential circumstantial evidence in favour of this.



    Probably not the main reason. I only stated that it's one factor. We don't have the appropriate ancient samples to model Slavic populations yet (particularly for East Slavs - Closest thing so far is CWC Baltic).
    Yes because the modelling fits are horrendous they seem to be missing a population either that or the sample quality was bad. Though I imagine rural Russians would be quite similar in particular Cossack or Old Timers you mentioned , I think their minor Uralic/ Okunevo admixture throws them off. I noticed the same with Pamiri Tajiks.

  8. The Following 3 Users Say Thank You to pegasus For This Useful Post:

     Agamemnon (01-09-2019),  DMXX (01-08-2019),  Hando (01-26-2019)

  9. #25
    Indefinite Hiatus
    Posts
    3,145
    Sex
    Y-DNA
    R2a*-M124 (L295-)
    mtDNA
    D4j5*

    England
    Quote Originally Posted by pegasus View Post
    Yes because the modelling fits are horrendous they seem to be missing a population either that or the sample quality was bad. Though I imagine rural Russians would be quite similar in particular Cossack or Old Timers you mentioned , I think their minor Uralic/ Okunevo admixture throws them off. I noticed the same with Pamiri Tajiks.
    Yeah, if I were to guess, we're probably missing a unique EHG-related population that existed after (and south of) Karelia_HG. Adding in EHG-rich groups in isolation of others seems to improve things by a pinch. That's probably a bigger factor than any sort of genetic drift.

  10. The Following 6 Users Say Thank You to DMXX For This Useful Post:

     Agamemnon (01-09-2019),  Eterne (01-08-2019),  Hando (01-26-2019),  pegasus (01-10-2019),  Rafe (01-11-2019),  Ryukendo (01-19-2019)

  11. #26
    Registered Users
    Posts
    28
    Sex

    Quote Originally Posted by Helves View Post
    I think when the relevant non-IE population, with a ratio of WHG-EHG-EEF that works for modern day Balto-Salvs is sampled we will see their Yamnaya levels drop probably just a bit from what the 2015 Haak et al study showed.
    Right now using Davids spearsheet I can't get any good fits at all with a Steppe-HG-EEF run.
    There is Baltic_BA from Latvia, for example, which is even more WHG(+EHG) shifted.
    In fact, it seems, that there is a sort of a cline from Latvian BA to Lithuanian BA to modern Balts to modern Northern Slavs. You get it by adding some extra Central-European admix to Baltic Bronze Age. Perhaps, we will find the source of this "Central European affinity" in Welzin_BA (Tollensse) and/or Wielbark samples

  12. The Following 2 Users Say Thank You to FenriR For This Useful Post:

     Coldmountains (01-08-2019),  Hando (01-26-2019)

  13. #27
    Registered Users
    Posts
    3,035
    Sex

    Quote Originally Posted by Helves View Post
    How can modern day Balto-Slavs rich in R1a-M417 not share the same amount of drift with the CWC samples the same way NW Euros do with the Bell Beakers?
    Because Balto-Slavic drift has a big effect on some of the G25 dimensions. NW Euros don't have nearly the same effect on the analysis, probably because of their higher effective population sizes.

    This is the trade off of producing an analysis that can model ancient ancestry but also distinguish many closely related populations from each other, like NE and NW Europeans.

    Another method, like qpAdm, which is based on formal stats, largely ignores recent drift, and so it has a hard time distinguishing groups with practically the same levels of the same ancient ancestral components. For instance, you'd have a very hard time telling apart Danes from Poles with qpAdm, but not with the G25.

    So you have to stop fixating about the distance discrepancies in the G25 models, and accept that they can't be compared directly for many ethnic groups. If you want to compare distances directly like this between ethnic groups, then you have to use qpAdm.

  14. The Following 13 Users Say Thank You to Generalissimo For This Useful Post:

     Agamemnon (01-09-2019),  Alain (01-09-2019),  Coldmountains (01-08-2019),  DMXX (01-08-2019),  Eihwaz (01-18-2019),  FenriR (01-09-2019),  Hando (01-26-2019),  Helves (01-09-2019),  Jessie (01-09-2019),  JMcB (01-10-2019),  Michał (01-09-2019),  Psynome (01-09-2019),  Ruderico (01-09-2019)

  15. #28
    Registered Users
    Posts
    6,549
    Sex
    Omitted
    Y-DNA
    L21
    mtDNA
    H

    At last - an interesting thread!

  16. The Following 3 Users Say Thank You to alan For This Useful Post:

     Hando (01-26-2019),  Rafe (01-11-2019),  sgdavies@hotmail.com (01-10-2019)

  17. #29
    Registered Users
    Posts
    46
    Sex

    Quote Originally Posted by DMXX View Post
    Yeah, if I were to guess, we're probably missing a unique EHG-related population that existed after (and south of) Karelia_HG. Adding in EHG-rich groups in isolation of others seems to improve things by a pinch. That's probably a bigger factor than any sort of genetic drift.
    How much of a contribution are you thinking this population made? When do you think they were absorbed so to speak?

  18. The Following User Says Thank You to xenus For This Useful Post:

     Hando (01-26-2019)

  19. #30
    Registered Users
    Posts
    604
    Sex
    Omitted

    How would non-Yamna part of Sintashta and of Baltic BA differ?
    I imagine Baltic BA is more towards WHG/EHG and Sintashta EEF?

  20. The Following User Says Thank You to parastais For This Useful Post:

     Hando (01-26-2019)

Page 3 of 7 FirstFirst 12345 ... LastLast

Similar Threads

  1. Replies: 17
    Last Post: 02-25-2019, 08:52 AM
  2. Steppe Pastoralists into Western Europe
    By A Norfolk L-M20 in forum Archaeology (Prehistory)
    Replies: 0
    Last Post: 12-23-2016, 11:50 AM
  3. New Leak: CWC=73% Yamna, modern North Euros=50% Yamna.
    By Krefter in forum Ancient (aDNA)
    Replies: 194
    Last Post: 02-12-2015, 03:10 AM
  4. Replies: 2
    Last Post: 01-09-2015, 06:57 AM
  5. Replies: 2
    Last Post: 09-26-2013, 11:51 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
  •