Eupedia Forums
Site NavigationEupedia Top > Eupedia Forum & Japan Forum
Page 3 of 3 FirstFirst 123
Results 51 to 61 of 61

Thread: Autosomal results of Neolithic genome from Iberia

  1. #51
    Advisor LeBrok's Avatar
    Join Date
    18-11-09
    Location
    Calgary
    Posts
    10,295

    Y-DNA haplogroup
    R1b Z2109
    MtDNA haplogroup
    H1c

    Ethnic group
    Citizen of the world
    Country: Canada-Alberta



    Quote Originally Posted by Alan View Post
    Let's take Tajiks as example.

    They have ~45% ENF. => https://docs.google.com/spreadsheets...t#gid=74932529

    But they have ~25% Atlantic_Med/Caucasus/Southwest Asian => https://docs.google.com/spreadsheet/...DcTZEYlE#gid=0

    20% ENF remain.

    So which of these remaining components are the most likely source for the rest of the ENF. We have to chose between North European, Gedrosia, South Asian or Siberian.

    I bet my money on Gedrosia. And we shouldn't forget the relation between Gedrosia and Caucasus is like the relation between Northwest European and Northeast European. So if Caucasus is predominantly ENF, so Gedrosia must have allot of ENF itself.

    And we also had already this discussion about low EEF. It isn't something new.

    The reason why the EEF is so low is because Yamna does not have the European farmer type ancestry. Their ENF ancestry is directly the highland pastoralist type. So EEF is unlikely to show up.
    There is at least 75% ENF in EEF. So if EEF doesn't show in Yamna it is very unlikely that there is substantial ENF in Yamna. I would be surprised if other components have more than 10% of ENF. If we find Gedrosia in Neolithic in Near East I will agree with you that it contained ENF. ;)
    Be wary of people who tend to glorify the past, underestimate the present, and demonize the future.

  2. #52
    Elite member
    Join Date
    14-11-10
    Posts
    2,508

    Y-DNA haplogroup
    R1a1a1
    MtDNA haplogroup
    HV2a1 +G13708A

    Ethnic group
    Kurdish
    Country: Germany



    Quote Originally Posted by LeBrok View Post
    There is at least 75% ENF in EEF. So if EEF doesn't show in Yamna it is very unlikely that there is substantial ENF in Yamna. I would be surprised if other components have more than 10% of ENF. If we find Gedrosia in Neolithic in Near East I will agree with you that it contained ENF. ;)
    Bedouins are 85%. Going by that, we need to assume Bedoins beeing ~40% Atlanto_Med. That is impossible.

    You see what I mean? Atlanto_Med came to existence with ENF(2/3) mixing with some WHG(1/3). But that doesn't mean every ENF group needs to have some Atlanto_Med, because this component is a specific type which was probably born in Europe because of additional WHG admixture.

    West Asian highlanders also score only half that much EEF as they score ENF, simply because they do not have this WHG admixed type of farmer ancestry. This is why you will not find any EEF in Yamna.

    Imagine if we had a population with 50% Northwest European component. And I would tell you they are not "North European" admixed because look here they show no Northeast European ancestry :)

    Thats exactly the problem here. Just because someone has no EEF('Northeast European') doesn't mean he has no ENF('North European'), because his ENF is possibly the highlander type('West Asian').


    LeBrock we know that Yamna is ~25% ENF but Yamna has only 7% Caucasus/Atlanto_Med. The rest must come from Gedrosia. Just like in my example with Tajiks there is no other way or component it could have come from. There is no denying there.
    And Even Reich said the ENF ancestry in Yamna is typical Near Eastern and came directly from the highlands and not of the type found in Europe(EEF).

    So everything adds up :)

    Believe me Gedrosia is roughly 60% ENF.
    Last edited by Alan; 09-03-15 at 17:22.

  3. #53
    Elite member
    Join Date
    25-10-11
    Location
    Brittany
    Age
    72
    Posts
    4,921

    Y-DNA haplogroup
    R1b - L21/S145*
    MtDNA haplogroup
    H3c

    Ethnic group
    more celtic
    Country: France



    Gedrosia of today in Near East and Pakistan and surroundings is not by force exactly the Gedrosia we find elsewhere more northern or we FOUND before -
    and even this southern Gedrosia we constate nowaday is not by force from a Near Eastern unique source - ANE seems new in Near Eastern (differences between sub-populations of ancient endogamy and the remnant) as Gedrosia is, even if Gedrosia is not equal ANE - a population showing ANE shew also Gedrosia (whatever a tiny link between both, uniquely statistical in some population, but this "cocktail" for me evocates a Central or South-Central Asian population, not a genuine old Near-Eastern one - just my point, trying to find my way in darkness (here I leave a tear dropping down from my eye)

  4. #54
    Regular Member
    Join Date
    23-03-12
    Posts
    32

    Y-DNA haplogroup
    R-L21
    MtDNA haplogroup
    U4c1

    Country: France



    Quote Originally Posted by Alan View Post
    So here it is, the results of the R1b1 sample from 5100 BC late neolithic Spain.

    K12b

    • 74.26% Atlantic_Med
    • 18.37% Caucasus
    • 3.70% Southwest_Asian
    • 3.67% Northwest_African
    • 0.00% East_African
    • 0.00% East_Asian
    • 0.00% Gedrosia
    • 0.00% North_European
    • 0.00% Siberian
    • 0.00% South_Asian
    • 0.00% Southeast_Asian
    • 0.00% Sub_Saharan



    Typical European farmer with the typical Mediterranean/Southwest Asian/Caucasus DNA + some North European from admixing with WHG (also typical for European farmers).
    This is quite irritating. We have Samara H&G with not much sign of ENF but yet we have another R1b with typical farmer DNA.

    .
    I do not agree!
    The Spanish neolithic is different from the Neolithic of Europe central as LBK.
    See the difference:

    LBK (F999916)
    #PopulationPercent
    1Atlantic_Med 54.922
    Caucasus30.33
    Southwest_Asian10.784
    Northwest_African3.795
    North_European0.146
    Southeast_Asian0.06

    The Spanish Neolithic have more Atlantic_Med and less of caucasus and SW_Asian.

    I do not know why everyone thinks Atlantic-Med comes from near-east!


    Only SW_Asian and caucasus were probably brought by farmers of Near-East

  5. #55
    Elite member
    Join Date
    21-01-14
    Posts
    533


    Country: UK - Wales



    Quote Originally Posted by gervais View Post
    I do not agree!
    The Spanish neolithic is different from the Neolithic of Europe central as LBK.
    See the difference:

    LBK (F999916)
    #PopulationPercent
    1Atlantic_Med 54.922
    Caucasus30.33
    Southwest_Asian10.784
    Northwest_African3.795
    North_European0.146
    Southeast_Asian0.06

    The Spanish Neolithic have more Atlantic_Med and less of caucasus and SW_Asian.

    I do not know why everyone thinks Atlantic-Med comes from near-east!


    Only SW_Asian and caucasus were probably brought by farmers of Near-East
    Quite. The Atlantic coast had a different climate and that led to a different story.

  6. #56
    Advisor Angela's Avatar
    Join Date
    02-01-11
    Posts
    19,255


    Ethnic group
    Italian
    Country: USA - New York



    2 members found this post helpful.
    Quote Originally Posted by gervais View Post
    I do not agree!
    The Spanish neolithic is different from the Neolithic of Europe central as LBK.
    See the difference:

    LBK (F999916)
    #PopulationPercent
    1Atlantic_Med 54.922
    Caucasus30.33
    Southwest_Asian10.784
    Northwest_African3.795
    North_European0.146
    Southeast_Asian0.06

    The Spanish Neolithic have more Atlantic_Med and less of caucasus and SW_Asian.

    I do not know why everyone thinks Atlantic-Med comes from near-east!


    Only SW_Asian and caucasus were probably brought by farmers of Near-East
    Atlantic Med is probably best described as a combination of Neolithic farmer alleles from the Near East and a minority WHG component. The creator of the calculator being used by you as well as by the poster Alan explained the relationship of these components here:
    http://dienekes.blogspot.com/2012/08...-k12b-and.html
    http://dienekes.blogspot.com/2012/09...decad-k7b.html

    The graphs are particularly interesting. Note that S.W. Asian is just a "specialized" form of "Caucasus", and "Caucasus" goes into "Atlantic Med".

    I wouldn't quarrel with Alan's estimate of about a 25% or so WHG percentage in Atlantic Med.

    It's important to realize that these calculators, while they were useful in their day, produce components which represent much more recent geographical "poolings" which are the result of many layers of migration. Formal stats are much more informative. For that, you have to read Lazaridis et al, and Haak et al.
    http://biorxiv.org/content/early/2013/12/23/001552
    http://biorxiv.org/content/early/2015/02/10/013433

    In Haak et al, the resnorm stats are particularly interesting. As you can see, the French are 69% Early Neolithic. (Of course, these numbers are the result not only of Neolithic migrations, but also of those of Bronze Age "Indo-Europeans".) It would be impossible to reach numbers like this if only S.W.Asian and Caucasus represented migration of peoples with ancestry from the Near East.


    Attachment 7147


    Non si fa il proprio dovere perchè qualcuno ci dica grazie, lo si fa per principio, per se stessi, per la propria dignità. Oriana Fallaci

  7. #57
    Regular Member
    Join Date
    23-03-12
    Posts
    32

    Y-DNA haplogroup
    R-L21
    MtDNA haplogroup
    U4c1

    Country: France



    Quote Originally Posted by Angela View Post
    Atlantic Med is probably best described as a combination of Neolithic farmer alleles from the Near East and a minority WHG component. The creator of the calculator being used by you as well as by the poster Alan explained the relationship of these components here:
    http://dienekes.blogspot.com/2012/08...-k12b-and.html
    http://dienekes.blogspot.com/2012/09...decad-k7b.html

    The graphs are particularly interesting. Note that S.W. Asian is just a "specialized" form of "Caucasus", and "Caucasus" goes into "Atlantic Med".

    I wouldn't quarrel with Alan's estimate of about a 25% or so WHG percentage in Atlantic Med.

    It's important to realize that these calculators, while they were useful in their day, produce components which represent much more recent geographical "poolings" which are the result of many layers of migration. Formal stats are much more informative. For that, you have to read Lazaridis et al, and Haak et al.
    http://biorxiv.org/content/early/2013/12/23/001552
    http://biorxiv.org/content/early/2015/02/10/013433

    In Haak et al, the resnorm stats are particularly interesting. As you can see, the French are 69% Early Neolithic. (Of course, these numbers are the result not only of Neolithic migrations, but also of those of Bronze Age "Indo-Europeans".) It would be impossible to reach numbers like this if only S.W.Asian and Caucasus represented migration of peoples with ancestry from the Near East.


    Attachment 7147
    Yes, the component closest to Atlantic-Med is Caucasus, but the closest component of North-European is Atlantic-med.
    How the Basques have 0% Caucasus (the only of Europe!!) and Kostenki14 a russian Palaeolithic (- 35,000) has 21.46% of Atlantic-Med!


    It is as if we were dealing with two migrations: Atlantic-Med and then caucasus.
    The question is: the two were farmers or only one of the two?

    Basque:
    Atlantic-Med: 73,1
    North-European: 17,1
    Gedrosia 9,8
    Caucasus: 0

    Kostenki14
    #PopulationPercent
    1North_European28.8
    2Atlantic_Med21.46
    3South_Asian15.7
    4Gedrosia12.38
    5Southeast_Asian6.12
    6Southwest_Asian4.95
    7East_African3.87
    8Siberian3.78
    9Northwest_African1.65
    10Sub_Saharan1.15
    11East_Asian0.15

  8. #58
    Regular Member
    Join Date
    21-01-12
    Posts
    13


    Ethnic group
    Northern european
    Country: France



    Quote Originally Posted by Angela View Post
    Atlantic Med is probably best described as a combination of Neolithic farmer alleles from the Near East and a minority WHG component. The creator of the calculator being used by you as well as by the poster Alan explained the relationship of these components here:
    http://dienekes.blogspot.com/2012/08...-k12b-and.html
    http://dienekes.blogspot.com/2012/09...decad-k7b.html

    The graphs are particularly interesting. Note that S.W. Asian is just a "specialized" form of "Caucasus", and "Caucasus" goes into "Atlantic Med".

    I wouldn't quarrel with Alan's estimate of about a 25% or so WHG percentage in Atlantic Med.

    It's important to realize that these calculators, while they were useful in their day, produce components which represent much more recent geographical "poolings" which are the result of many layers of migration. Formal stats are much more informative. For that, you have to read Lazaridis et al, and Haak et al.
    http://biorxiv.org/content/early/2013/12/23/001552
    http://biorxiv.org/content/early/2015/02/10/013433

    In Haak et al, the resnorm stats are particularly interesting. As you can see, the French are 69% Early Neolithic. (Of course, these numbers are the result not only of Neolithic migrations, but also of those of Bronze Age "Indo-Europeans".) It would be impossible to reach numbers like this if only S.W.Asian and Caucasus represented migration of peoples with ancestry from the Near East.


    Attachment 7147
    In the last paper of Haak, french are 43 % EN ( not 69%) 57 % WHG
    (figure S9.24 p.120)


    That is maybe different in the paper of 2013 ?


    But I agree with Gervais, there is a difference between the Atlantic Neolithics and those of the Bassin Danunian, the Caucasian component is higher in the LBKs

  9. #59
    Advisor Angela's Avatar
    Join Date
    02-01-11
    Posts
    19,255


    Ethnic group
    Italian
    Country: USA - New York



    Quote Originally Posted by Armoricain View Post
    In the last paper of Haak, french are 43 % EN ( not 69%) 57 % WHG
    (figure S9.24 p.120)


    That is maybe different in the paper of 2013 ?


    But I agree with Gervais, there is a difference between the Atlantic Neolithics and those of the Bassin Danunian, the Caucasian component is higher in the LBKs
    I'm aware of the figure to which you referred. Haak et al did exhaustive modeling. With each successive model they tried to reduce the residuals (or improve the fit) more and more. The model to which you referred only used EN and WHG. In the subsequent model (page 121), they added the Yamnaya, and the result for EN in the French was 51.2, with lower residuals. In the next model, the figure was 64.3. In the figure to which I linked but which did not post, Figure S9.27 on page 124 of the Haak et al 2015 supplement, the figure for EN is indeed 69%. I'm going to try to post it again.

    Lazaridis and Haak resnorm table.JPG

    Whatever the figure you want to use, whether it's the one on page 121 of 51.2 percent, or even the 43% EN from a model with high residuals (i.e. not a good fit), it doesn't change the fact that based on Dodecad K12b, the total of S.W.Asian and Caucasus in the French is 11%. Obviously, there is EN in "Atlantic Med" as well, and as Dienekes pointed out about his own calculator, in North European.

    As to your last statement, I personally don't think it is very helpful to rely on the once useful but obviously flawed admixture calculators based on modern populations and relatively modern geographical groupings or components, when the formal stats based on actual ancient genomes give far more accurate results. (I concede you don't get the tightest possible fit using only ancient genomes, as the authors realized, which is why they added the Nganasan and the Bedouin to the graphic I posted.)

    In so far as I can tell, PCA's and formal stats and sophisticated Admixture programs indicate that the early EEF samples are all pretty much alike, which is exactly what this Lab has said numerous times. They certainly cluser closely together on PCA's, and there is no marked "eastward" shift in LBK samples:

    See Gamba et al:
    http://www.nature.com/ncomms/2014/14...comms6257.html


    It's true that the LBK did not at first mix very much. As time passed, there was more admixture, but in terms of the initial farmers, whether they were "Cardial" or "LBK", I haven't seen anything which would indicate major differences. If you have data to that effect which is persuasive, i.e. formal stats from academics, not Blogger calculators, I would of course change my opinion. It's true, of course that the EN which fed into Europe through Yamnaya is more "Caucasus" like.

    You might also want to take a look at Paschou et al and the thread here discussing it.

    http://www.pnas.org/content/111/25/9211.abstract

    http://www.pnas.org/content/suppl/20...11111.sapp.pdf

    http://www.eupedia.com/forum/threads...hlight=Paschou

  10. #60
    Elite member
    Join Date
    21-01-14
    Posts
    533


    Country: UK - Wales



    in terms of the initial farmers, whether they were "Cardial" or "LBK", I haven't seen anything which would indicate major differences.
    I don't think there's going to be major differences because they're all mixed to a large extent but the *small* differences will be what provides the clues to the detailed history imo.

  11. #61
    Regular Member
    Join Date
    21-01-12
    Posts
    13


    Ethnic group
    Northern european
    Country: France



    Quote Originally Posted by Angela View Post
    I'm aware of the figure to which you referred. Haak et al did exhaustive modeling. With each successive model they tried to reduce the residuals (or improve the fit) more and more. The model to which you referred only used EN and WHG. In the subsequent model (page 121), they added the Yamnaya, and the result for EN in the French was 51.2, with lower residuals. In the next model, the figure was 64.3. In the figure to which I linked but which did not post, Figure S9.27 on page 124 of the Haak et al 2015 supplement, the figure for EN is indeed 69%. I'm going to try to post it again.

    Lazaridis and Haak resnorm table.JPG

    Whatever the figure you want to use, whether it's the one on page 121 of 51.2 percent, or even the 43% EN from a model with high residuals (i.e. not a good fit), it doesn't change the fact that based on Dodecad K12b, the total of S.W.Asian and Caucasus in the French is 11%. Obviously, there is EN in "Atlantic Med" as well, and as Dienekes pointed out about his own calculator, in North European.

    As to your last statement, I personally don't think it is very helpful to rely on the once useful but obviously flawed admixture calculators based on modern populations and relatively modern geographical groupings or components, when the formal stats based on actual ancient genomes give far more accurate results. (I concede you don't get the tightest possible fit using only ancient genomes, as the authors realized, which is why they added the Nganasan and the Bedouin to the graphic I posted.)

    In so far as I can tell, PCA's and formal stats and sophisticated Admixture programs indicate that the early EEF samples are all pretty much alike, which is exactly what this Lab has said numerous times. They certainly cluser closely together on PCA's, and there is no marked "eastward" shift in LBK samples:

    See Gamba et al:
    http://www.nature.com/ncomms/2014/14...comms6257.html


    It's true that the LBK did not at first mix very much. As time passed, there was more admixture, but in terms of the initial farmers, whether they were "Cardial" or "LBK", I haven't seen anything which would indicate major differences. If you have data to that effect which is persuasive, i.e. formal stats from academics, not Blogger calculators, I would of course change my opinion. It's true, of course that the EN which fed into Europe through Yamnaya is more "Caucasus" like.

    You might also want to take a look at Paschou et al and the thread here discussing it.

    http://www.pnas.org/content/111/25/9211.abstract

    http://www.pnas.org/content/suppl/20...11111.sapp.pdf

    http://www.eupedia.com/forum/threads...hlight=Paschou
    Thank you for all your explanations , I will consult your links with great interest.

Page 3 of 3 FirstFirst 123

Posting Permissions

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