The Unz Review - Mobile
A Collection of Interesting, Important, and Controversial Perspectives Largely Excluded from the American Mainstream Media
 TeasersGene Expression Blog
Why I Still Lean Toward a Sub-Saharan African Origin for Modern Humanity
🔊 Listen RSS
Email This Page to Someone

 Remember My Information



=>

Bookmark Toggle AllToCAdd to LibraryRemove from Library • BShow CommentNext New CommentNext New Reply
Search Text Case Sensitive  Exact Words  Include Comments
List of Bookmarks

Dienekes argues:

Yes, it’s possible that the divergence event happened in Africa. It’s also possible that it happened in Asia. I see no reason to prefer one or the other. After all, the main piece of evidence in favor of Out-of-Africa is that Eurasians are nested within African genetic variation.

If this paper is right, this is no longer the case: Africans are nested in Eurasian variation as a whole, inclusive of both modern Eurasians and the mystery population.

I’m not sure that this is the main piece, though it was a major reason. Here are two other reasons:

1) Anatomically modern humans show up in Africa first.

2) The deepest divergences in the mtDNA, Y chromosomes, and autosomes, are all found within Africa (in particular, between hunter-gatherer African populations and everyone else).

Let’s consider the alternative models in terms of the genetics.

Near Eastern origin of modern humanity: Rapid expansion of humans out of this area. The deep divergences within Africa are happenstance; all the deep lineages outside of Africa were replaced by the “Out of Africa” population.

African origin of modern humanity: Rapid expansion of “Out of Africa” humans 60,000 years ago, with deep structure within Africa that dates back to 200,000 years ago.

Both views match the data. But the latter model seems more parsimonious to me. And as I stated above I’m to understand it is more well aligned with the archaeology. But time will tell. The first view may be true.

 
• Category: Science • Tags: Genetics, Genomics 
Hide 25 CommentsLeave a Comment
Commenters to Ignore...to FollowEndorsed Only
Trim Comments?
  1. The most deplorable one [AKA "Fourth doorman of the apocalypse"] says:

    I am comfortable with the notion that Homo Sapiens Sapiens first arose in Africa and that on its journey out of Africa Homo Sapiens Sapiens acquired genes from other archaic Homo species.

    I am also comfortable with the notion that modern Sub-Saharan Africans are derived.

    Read More
    ReplyAgree/Disagree/Etc. More... This Commenter Display All Comments
  2. I’m no longer sure of the meaning of the question at this point. Africans ARE nested in European variation; Europeans are nested in African variation. Some African lineages have had relatively little contact with non-African lineages in 100-200,000 years. Eurasian lineages have ancestry from Neanderthals that seem not to have otherwise had contact with African genomes for 600,000 years. Yet, the variation in phenotype in all these populations is for the most part nested within the variation for any other population, certainly by comparison with any other animal. We are all virtually hairless, uniformly bipedal, talkative, social intelligent. We all have chins with a few exceptions who hide their disability with beards. It seems clear (and I’m not trying to lecture, you basically said as much in your previous post, and said it better) that humanization was an evolutionary process, not a Gouldian punctuated moment followed by equilibrium, and that all strands of hominid development have been subject to the dynamics of that process for millennia on millennia.

    Read More
    • Replies: @Razib Khan
    how are africans nested in european variation? don't follow.
    ReplyAgree/Disagree/Etc. More... This Commenter This Thread Hide Thread Display All Comments
  3. It’s a pity the divergence of the ghost population is so close to San’s, can’t see who is nested within what.

    Modern Africans perhaps within ancient Eurasian variation (which is still within ancient African) and modern Eurasians within modern African. Unless certain minor oddities like the admixture edges from Africa to East Asia in TreeMix are remnants of this first OOA, no evidence it left remnants anywhere so far.

    Read More
    ReplyAgree/Disagree/Etc. More... This Commenter Display All Comments
  4. Anonymous • Disclaimer says:

    A plausible explanation for the apparent diversity of African genomes is interbreeding between Eurasian immigrant sapiens and the mysterious, hypothesised ‘hominid x’, which is known for sure to account for a large portion of SS Africa genetics.
    The implication being that ‘hominid x’ persisted in the African continent unchanged and in large numbers for a truly vast stretch of time.

    Read More
    • Replies: @Razib Khan
    If it was plausible I would have mentioned it. There are reasons that doesn't work.
    , @Rick
    The main arguments against this are the Y and mt haplogroup trees.

    As far as we know, 99.99% (with the possible exception of the A00 Y chromosome) of modern humans do not have suspected archaic Y or mt haplogroups, and the trees for these largely match the topology of the trees for the bulk of the autosomal genes jn the same populations.

    Clearly (as seen in Eurasia) these can easily be lost due to drift or selection when <10% of the population has them.

    But, if most of the diversity of autosomes in Africa was due to admixture with various archaic humans, then there would be random deeply divergent haplogroups.

    Also, the average autosome divergence would be much more random and not correlated with the haplogroups at all.
    ReplyAgree/Disagree/Etc. More... This Commenter This Thread Hide Thread Display All Comments
  5. The recent Paleolithic Euro mtDNA paper refined the mitochondrial mutation rate. The upper bound age estimate for Eurasian M and N is now 55 kya. Overlapping the age of major Eurasian Y chromosomal haplogroups C, F and D. And we have a proto-Eurasian genome from 45 kya whose Neanderthal admixture dates to this “recent OOA” period.

    Previously, the Skhul and Qafzeh AMHs were thought to have died out since the fossil record shows Neanderthals repopulating West Asia. Now, we find AMH admixture in Neanderthals that is unrelated to modern Eurasians (in fact, modeling San or Yoruba as the source works slightly better).

    I think Dienekes has to approach Dziebel levels of mental gymnastics to suggest these results point to anything other than recent OOA.

    Read More
    • Replies: @ohwilleke
    One of the biggest open questions in the genetic history of modern humans, in my view, is whether there is continuity between Skhul and Qafzeh AMHs or related populations that left Africa around then, or alternatively, if that was an "Out of Africa that failed."

    Keep in my the the datas you are referring to with regard to mtDNA M and N, and with Y-DNA C, F and D, is "the most recent common ancestor" date. Those most recent common ancestor dates can't rule out the possibility that people with the mtDNA clades of L3 that we now call M and N , and that the Y-DNA clades of Y-DNA B that we call C and F (and other sister clades of mtDNA L3 and Y-DNA B each that didn't survive a bottleneck) for many thousands of years before the TMRCA data.

    In general, lineages created by mutations are rarely lost when populations are expanding, are overwhelmingly lost and reduced during periods of declining population, and are pruned more swiftly than your intuition would expect during period of stable population. If you have an initial Out of Africa population that isn't huge to start with and declines between 125kya and 55kya or so, before suddenly expanding again, it is not very easy to distinguish this population based upon genetic data alone from one that leaves Africa in the first instance at 55 kya with mutation rate based dating.

    Linkage disequilibrium could easily distinguish the two scenarios, but this method doesn't have sufficient time depth to distinguish the scenarios in contemporary genomes, although LD methods applied to old Upper Paleolithic whole genomes might be able to make that kind of distinction.

    I personally favor a scenario with an initial Out of Africa ca. 125kya that survives in South Asian and Arabian interior populations for a prolonged period (possibly going extinct in Arabia), expands from South Asia to SE Asia and beyond ca. 75 kya following the Toba exclusion, and has a secondary expansion from South Asia (and possibly also Arabia) ca. 55kya in all directions (including SE Asia, Central Asia, Australia and back migration evidenced by mtDNA U6 and M1 in Africa) except for a Neanderthal barrier in Europe that persists until ca. 40kya with volcanic induced climate change gives AMHs a momentary edge that leads to Neanderthal extinction or near extinction.

    I think that the handful of older purported AMH fossils in China are probably either misdated or misclassified.

    , @Razib Khan
    btw, the last statement was a low blow! ;-) i hope you're kidding.
    ReplyAgree/Disagree/Etc. More... This Commenter This Thread Hide Thread Display All Comments
  6. @Anonymous
    A plausible explanation for the apparent diversity of African genomes is interbreeding between Eurasian immigrant sapiens and the mysterious, hypothesised 'hominid x', which is known for sure to account for a large portion of SS Africa genetics.
    The implication being that 'hominid x' persisted in the African continent unchanged and in large numbers for a truly vast stretch of time.

    If it was plausible I would have mentioned it. There are reasons that doesn’t work.

    Read More
    ReplyAgree/Disagree/Etc. More... This Commenter This Thread Hide Thread Display All Comments
  7. Re: fossils, the Omo skull fromAfrica is the earliest right (195,000 BP)? I wonder if the Dali Man skull in China (around 200,000 BP) and the Jinniushan Man (also 200,000 BP?) which apparenyl have some modern human features*, and have been argued to be like archaic Homo Sapiens, will see reassessment with the introgression to Altai. Are we more likely to see them as representatives of an Southern Asian species of archaic H. Sap, now, or if too divergent an introgression from early H. Sap into other species…?

    *IIUC overlapping cranial volumes and shape features, like a vertically short and not very projecting face, which is otherwise an AMH unique feature.

    Read More
    ReplyAgree/Disagree/Etc. More... This Commenter Display All Comments
  8. @Anonymous
    A plausible explanation for the apparent diversity of African genomes is interbreeding between Eurasian immigrant sapiens and the mysterious, hypothesised 'hominid x', which is known for sure to account for a large portion of SS Africa genetics.
    The implication being that 'hominid x' persisted in the African continent unchanged and in large numbers for a truly vast stretch of time.

    The main arguments against this are the Y and mt haplogroup trees.

    As far as we know, 99.99% (with the possible exception of the A00 Y chromosome) of modern humans do not have suspected archaic Y or mt haplogroups, and the trees for these largely match the topology of the trees for the bulk of the autosomal genes jn the same populations.

    Clearly (as seen in Eurasia) these can easily be lost due to drift or selection when <10% of the population has them.

    But, if most of the diversity of autosomes in Africa was due to admixture with various archaic humans, then there would be random deeply divergent haplogroups.

    Also, the average autosome divergence would be much more random and not correlated with the haplogroups at all.

    Read More
    ReplyAgree/Disagree/Etc. More... This Commenter This Thread Hide Thread Display All Comments
  9. The model of Near Eastern origin is obviously correct because Genesis ;)

    Read More
    ReplyAgree/Disagree/Etc. More... This Commenter Display All Comments
  10. I Still Lean Toward a Sub-Saharan African Origin for Modern Humanity

    I have the same tendency based on current evidences. I favor the process is result of asymmetrical human gene flows.

    Read More
    ReplyAgree/Disagree/Etc. More... This Commenter Display All Comments
  11. I would think that if modern humans originated in the Near East the admixture with Neanderthals would have been more wide spread.

    In fact, I’d argue that the limited spread of modern human admixture into Eurasian archaics is actually further evidence for a sub-Saharan origin of modern humans. Clearly the barriers between human species and subspecies in Eurasia were not large. If modern humans’ presence in Eurasia was long lived and widespread, why was there no modern human admixture in European Neanderthals or in Denisovans? Neanderthals and Denisovans certainly had no issues admixing with one another – the conflicting trees based on mitochondrial and autosomal DNA is pretty good evidence for that, as well as the known admixture between the two.

    My question would be – what does this tell us about the distribution of modern humans in Africa 100,000-250,000 years ago? Presumably these modern humans would have crossed into the Middle East during the Abbassia Pluvial 120,000-90,000 years ago, but diverge from other surviving human populations prior to that. Does that tell us anything about where the ultimate cradle of modern human evolution was?

    Read More
    • Replies: @Karl Zimmerman
    The problem here is the converse. We're now fairly sure there were archaic populations in Africa. We're also pretty sure that they interbred with at least some Sub-Saharan Africans not only after Africans had split off from non-Africans, but after those African populations themselves had been diverging for tens of thousands of years. Using the example of the pygmies from the other post, we are talking about AMH potentially living essentially alongside archaics for 170,00 years before a major admixture episode. There were not substantial geographic or climactic barriers across Sub-Saharan Africa over this entire time period. How could we have first diverged from the archaics, and then managed not to replace them (by whatever means) over such a long period.

    This is the crux of one of Dienekes arguments at least. If it wasn't for the basal haplogroup data being rooted so firmly in Africa, I'd find the argument that the foundation of AMH was somewhere else - somewhere with minimal influence by archaics - was plausible. But aside from A00, I don't find it likely any human haplogroups introgressed.

    On a broader note, these new papers have made me think about what exactly "AMH" means in discussing the human genotype. As an example, if you found a population like the Khulwilm paper - which split off genetically before the common ancestor of all modern groups, and left no living descendants - is it right to call it AMH? I mean, it is closer related to us than to Neandertals by descent, but Neandertals contributed to the modern human genome - this population did not. As archeogenetics gets to be more sophisticated, I expect we'll have more and more tough questions like this, since clearly applying the species concept, let alone such a fuzzy concept as "anatomical modernity" - to recent human evolution is by nature arbitrary.

    , @Razib Khan
    Does that tell us anything about where the ultimate cradle of modern human evolution was?

    i think only ancient DNA can resolve this. i have a broad belief that it was SS-african, but weak confidence of east vs. south (for example). and it could have been n. africa or near eastern. just lower probability IMO. though i'm pretty sure german dziebel is wrong.
    ReplyAgree/Disagree/Etc. More... This Commenter This Thread Hide Thread Display All Comments
  12. @CupOfCanada
    I would think that if modern humans originated in the Near East the admixture with Neanderthals would have been more wide spread.

    In fact, I'd argue that the limited spread of modern human admixture into Eurasian archaics is actually further evidence for a sub-Saharan origin of modern humans. Clearly the barriers between human species and subspecies in Eurasia were not large. If modern humans' presence in Eurasia was long lived and widespread, why was there no modern human admixture in European Neanderthals or in Denisovans? Neanderthals and Denisovans certainly had no issues admixing with one another - the conflicting trees based on mitochondrial and autosomal DNA is pretty good evidence for that, as well as the known admixture between the two.

    My question would be - what does this tell us about the distribution of modern humans in Africa 100,000-250,000 years ago? Presumably these modern humans would have crossed into the Middle East during the Abbassia Pluvial 120,000-90,000 years ago, but diverge from other surviving human populations prior to that. Does that tell us anything about where the ultimate cradle of modern human evolution was?

    The problem here is the converse. We’re now fairly sure there were archaic populations in Africa. We’re also pretty sure that they interbred with at least some Sub-Saharan Africans not only after Africans had split off from non-Africans, but after those African populations themselves had been diverging for tens of thousands of years. Using the example of the pygmies from the other post, we are talking about AMH potentially living essentially alongside archaics for 170,00 years before a major admixture episode. There were not substantial geographic or climactic barriers across Sub-Saharan Africa over this entire time period. How could we have first diverged from the archaics, and then managed not to replace them (by whatever means) over such a long period.

    This is the crux of one of Dienekes arguments at least. If it wasn’t for the basal haplogroup data being rooted so firmly in Africa, I’d find the argument that the foundation of AMH was somewhere else – somewhere with minimal influence by archaics – was plausible. But aside from A00, I don’t find it likely any human haplogroups introgressed.

    On a broader note, these new papers have made me think about what exactly “AMH” means in discussing the human genotype. As an example, if you found a population like the Khulwilm paper – which split off genetically before the common ancestor of all modern groups, and left no living descendants – is it right to call it AMH? I mean, it is closer related to us than to Neandertals by descent, but Neandertals contributed to the modern human genome – this population did not. As archeogenetics gets to be more sophisticated, I expect we’ll have more and more tough questions like this, since clearly applying the species concept, let alone such a fuzzy concept as “anatomical modernity” – to recent human evolution is by nature arbitrary.

    Read More
    • Replies: @Razib Khan
    As archeogenetics gets to be more sophisticated, I expect we’ll have more and more tough questions like this, since clearly applying the species concept, let alone such a fuzzy concept as “anatomical modernity” – to recent human evolution is by nature arbitrary.

    i agree.
    , @CupOfCanada

    The problem here is the converse. We’re now fairly sure there were archaic populations in Africa. We’re also pretty sure that they interbred with at least some Sub-Saharan Africans not only after Africans had split off from non-Africans, but after those African populations themselves had been diverging for tens of thousands of years. Using the example of the pygmies from the other post, we are talking about AMH potentially living essentially alongside archaics for 170,00 years before a major admixture episode. There were not substantial geographic or climactic barriers across Sub-Saharan Africa over this entire time period. How could we have first diverged from the archaics, and then managed not to replace them (by whatever means) over such a long period.
     
    We probably did interbreed with them. We just don't have ancient DNA to reference. If we were a set of constantly interbreeding populations we may be somewhat blind to this admixture I think? The Pigmy groups may have just encountered the most highly diverged/late persisting group so the signal sticks out strongest. If we never separated from some other groups in the first place... yah.
    ReplyAgree/Disagree/Etc. More... This Commenter This Thread Hide Thread Display All Comments
  13. @CupOfCanada
    I would think that if modern humans originated in the Near East the admixture with Neanderthals would have been more wide spread.

    In fact, I'd argue that the limited spread of modern human admixture into Eurasian archaics is actually further evidence for a sub-Saharan origin of modern humans. Clearly the barriers between human species and subspecies in Eurasia were not large. If modern humans' presence in Eurasia was long lived and widespread, why was there no modern human admixture in European Neanderthals or in Denisovans? Neanderthals and Denisovans certainly had no issues admixing with one another - the conflicting trees based on mitochondrial and autosomal DNA is pretty good evidence for that, as well as the known admixture between the two.

    My question would be - what does this tell us about the distribution of modern humans in Africa 100,000-250,000 years ago? Presumably these modern humans would have crossed into the Middle East during the Abbassia Pluvial 120,000-90,000 years ago, but diverge from other surviving human populations prior to that. Does that tell us anything about where the ultimate cradle of modern human evolution was?

    Does that tell us anything about where the ultimate cradle of modern human evolution was?

    i think only ancient DNA can resolve this. i have a broad belief that it was SS-african, but weak confidence of east vs. south (for example). and it could have been n. africa or near eastern. just lower probability IMO. though i’m pretty sure german dziebel is wrong.

    Read More
    • Replies: @CupOfCanada
    If it can resolve it at all. IIRC you were leaning towards a multiregional-in-Africa origin for humans, right? Do you still lean that way?
    ReplyAgree/Disagree/Etc. More... This Commenter This Thread Hide Thread Display All Comments
  14. @Karl Zimmerman
    The problem here is the converse. We're now fairly sure there were archaic populations in Africa. We're also pretty sure that they interbred with at least some Sub-Saharan Africans not only after Africans had split off from non-Africans, but after those African populations themselves had been diverging for tens of thousands of years. Using the example of the pygmies from the other post, we are talking about AMH potentially living essentially alongside archaics for 170,00 years before a major admixture episode. There were not substantial geographic or climactic barriers across Sub-Saharan Africa over this entire time period. How could we have first diverged from the archaics, and then managed not to replace them (by whatever means) over such a long period.

    This is the crux of one of Dienekes arguments at least. If it wasn't for the basal haplogroup data being rooted so firmly in Africa, I'd find the argument that the foundation of AMH was somewhere else - somewhere with minimal influence by archaics - was plausible. But aside from A00, I don't find it likely any human haplogroups introgressed.

    On a broader note, these new papers have made me think about what exactly "AMH" means in discussing the human genotype. As an example, if you found a population like the Khulwilm paper - which split off genetically before the common ancestor of all modern groups, and left no living descendants - is it right to call it AMH? I mean, it is closer related to us than to Neandertals by descent, but Neandertals contributed to the modern human genome - this population did not. As archeogenetics gets to be more sophisticated, I expect we'll have more and more tough questions like this, since clearly applying the species concept, let alone such a fuzzy concept as "anatomical modernity" - to recent human evolution is by nature arbitrary.

    As archeogenetics gets to be more sophisticated, I expect we’ll have more and more tough questions like this, since clearly applying the species concept, let alone such a fuzzy concept as “anatomical modernity” – to recent human evolution is by nature arbitrary.

    i agree.

    Read More
    ReplyAgree/Disagree/Etc. More... This Commenter This Thread Hide Thread Display All Comments
  15. fwiw, the reason i dismiss lots of arcahic admixture as explaining african autosomal diversity is that you’d see a particular distribution of TMRCas of LOTS AN LOTS of haplotypes. imagine, for example, that 30% of khoisan ancestry is “archaic” and diverged ~500,000 years ago. 70% is “modern” and diverged 200,000 years ago. you’d have two clusters of gene phylogenies with different median coalescence dates to non-african humans, as the latter “modern” haplotypes segregating in the genome would merge more quickly to the last common ancestry than the large minority of “archaic” haplotypes. or, more visually, the haplotype trees would have two clusters across many genes. you don’t see this to my knowledge.

    Read More
    • Replies: @CupOfCanada
    If populations weren't well segregated, would it show this clustering? IE if there was a genetic cline of various types of humans across Africa, and the San accreted much of this cline as they moved south (assuming an origin outside of south Africa).

    Does the bulk of our DNA have TMRCAs of ~200,000 years? Is there anywhere to see a distribution of TMRCAs for various haplotypes across the genome as a whole?
    ReplyAgree/Disagree/Etc. More... This Commenter This Thread Hide Thread Display All Comments
  16. @Lank
    The recent Paleolithic Euro mtDNA paper refined the mitochondrial mutation rate. The upper bound age estimate for Eurasian M and N is now 55 kya. Overlapping the age of major Eurasian Y chromosomal haplogroups C, F and D. And we have a proto-Eurasian genome from 45 kya whose Neanderthal admixture dates to this "recent OOA" period.

    Previously, the Skhul and Qafzeh AMHs were thought to have died out since the fossil record shows Neanderthals repopulating West Asia. Now, we find AMH admixture in Neanderthals that is unrelated to modern Eurasians (in fact, modeling San or Yoruba as the source works slightly better).

    I think Dienekes has to approach Dziebel levels of mental gymnastics to suggest these results point to anything other than recent OOA.

    One of the biggest open questions in the genetic history of modern humans, in my view, is whether there is continuity between Skhul and Qafzeh AMHs or related populations that left Africa around then, or alternatively, if that was an “Out of Africa that failed.”

    Keep in my the the datas you are referring to with regard to mtDNA M and N, and with Y-DNA C, F and D, is “the most recent common ancestor” date. Those most recent common ancestor dates can’t rule out the possibility that people with the mtDNA clades of L3 that we now call M and N , and that the Y-DNA clades of Y-DNA B that we call C and F (and other sister clades of mtDNA L3 and Y-DNA B each that didn’t survive a bottleneck) for many thousands of years before the TMRCA data.

    In general, lineages created by mutations are rarely lost when populations are expanding, are overwhelmingly lost and reduced during periods of declining population, and are pruned more swiftly than your intuition would expect during period of stable population. If you have an initial Out of Africa population that isn’t huge to start with and declines between 125kya and 55kya or so, before suddenly expanding again, it is not very easy to distinguish this population based upon genetic data alone from one that leaves Africa in the first instance at 55 kya with mutation rate based dating.

    Linkage disequilibrium could easily distinguish the two scenarios, but this method doesn’t have sufficient time depth to distinguish the scenarios in contemporary genomes, although LD methods applied to old Upper Paleolithic whole genomes might be able to make that kind of distinction.

    I personally favor a scenario with an initial Out of Africa ca. 125kya that survives in South Asian and Arabian interior populations for a prolonged period (possibly going extinct in Arabia), expands from South Asia to SE Asia and beyond ca. 75 kya following the Toba exclusion, and has a secondary expansion from South Asia (and possibly also Arabia) ca. 55kya in all directions (including SE Asia, Central Asia, Australia and back migration evidenced by mtDNA U6 and M1 in Africa) except for a Neanderthal barrier in Europe that persists until ca. 40kya with volcanic induced climate change gives AMHs a momentary edge that leads to Neanderthal extinction or near extinction.

    I think that the handful of older purported AMH fossils in China are probably either misdated or misclassified.

    Read More
    • Replies: @Razib Khan
    most of the data point to recent diversification of almost all ancestry outside of africa. also, the neanderthal admixture in non-africans is pretty clearly around 50,000 years BP. if there was an older out of africa, it's genetically signal is being obscured by the 50 K BP event.
    , @Lank
    Early OOA as the source of modern Eurasians does not work without substantial back-migration into Africa as well (not just Y-DNA CT but BT as well, not just mtDNA L3 but mtDNA L3'4, perhaps L2 as well), considering the more recent shared lineages. This is still possible, but it's hardly the most parsimonious scenario.

    The recent paper dates mtDNA M and N to 47-55 kya, which should make L3 about 60-65 ky old. The branches from the L3 node are not that long.

    C, F, and D are somewhat longer branches from CF/DE (almost 20 ky removed per YFull). But if you look at the downstream subclades, C/F/D start bifurcating like crazy 48 ky, so interestingly pretty much simultaneously. This seems to correspond with the rapid Eurasian expansion, with the long branches perhaps being a result of the Eurasian bottleneck. Interestingly, African-affiliated E starts to split up a bit earlier than the other CT groups, around 54 kya.

    IMHO, this pattern would be consistent with an OOA around 55-60 kya, followed by a rapid expansion.
    ReplyAgree/Disagree/Etc. More... This Commenter This Thread Hide Thread Display All Comments
  17. @ohwilleke
    One of the biggest open questions in the genetic history of modern humans, in my view, is whether there is continuity between Skhul and Qafzeh AMHs or related populations that left Africa around then, or alternatively, if that was an "Out of Africa that failed."

    Keep in my the the datas you are referring to with regard to mtDNA M and N, and with Y-DNA C, F and D, is "the most recent common ancestor" date. Those most recent common ancestor dates can't rule out the possibility that people with the mtDNA clades of L3 that we now call M and N , and that the Y-DNA clades of Y-DNA B that we call C and F (and other sister clades of mtDNA L3 and Y-DNA B each that didn't survive a bottleneck) for many thousands of years before the TMRCA data.

    In general, lineages created by mutations are rarely lost when populations are expanding, are overwhelmingly lost and reduced during periods of declining population, and are pruned more swiftly than your intuition would expect during period of stable population. If you have an initial Out of Africa population that isn't huge to start with and declines between 125kya and 55kya or so, before suddenly expanding again, it is not very easy to distinguish this population based upon genetic data alone from one that leaves Africa in the first instance at 55 kya with mutation rate based dating.

    Linkage disequilibrium could easily distinguish the two scenarios, but this method doesn't have sufficient time depth to distinguish the scenarios in contemporary genomes, although LD methods applied to old Upper Paleolithic whole genomes might be able to make that kind of distinction.

    I personally favor a scenario with an initial Out of Africa ca. 125kya that survives in South Asian and Arabian interior populations for a prolonged period (possibly going extinct in Arabia), expands from South Asia to SE Asia and beyond ca. 75 kya following the Toba exclusion, and has a secondary expansion from South Asia (and possibly also Arabia) ca. 55kya in all directions (including SE Asia, Central Asia, Australia and back migration evidenced by mtDNA U6 and M1 in Africa) except for a Neanderthal barrier in Europe that persists until ca. 40kya with volcanic induced climate change gives AMHs a momentary edge that leads to Neanderthal extinction or near extinction.

    I think that the handful of older purported AMH fossils in China are probably either misdated or misclassified.

    most of the data point to recent diversification of almost all ancestry outside of africa. also, the neanderthal admixture in non-africans is pretty clearly around 50,000 years BP. if there was an older out of africa, it’s genetically signal is being obscured by the 50 K BP event.

    Read More
    ReplyAgree/Disagree/Etc. More... This Commenter This Thread Hide Thread Display All Comments
  18. @ryanwc
    I'm no longer sure of the meaning of the question at this point. Africans ARE nested in European variation; Europeans are nested in African variation. Some African lineages have had relatively little contact with non-African lineages in 100-200,000 years. Eurasian lineages have ancestry from Neanderthals that seem not to have otherwise had contact with African genomes for 600,000 years. Yet, the variation in phenotype in all these populations is for the most part nested within the variation for any other population, certainly by comparison with any other animal. We are all virtually hairless, uniformly bipedal, talkative, social intelligent. We all have chins with a few exceptions who hide their disability with beards. It seems clear (and I'm not trying to lecture, you basically said as much in your previous post, and said it better) that humanization was an evolutionary process, not a Gouldian punctuated moment followed by equilibrium, and that all strands of hominid development have been subject to the dynamics of that process for millennia on millennia.

    how are africans nested in european variation? don’t follow.

    Read More
    ReplyAgree/Disagree/Etc. More... This Commenter This Thread Hide Thread Display All Comments
  19. @Lank
    The recent Paleolithic Euro mtDNA paper refined the mitochondrial mutation rate. The upper bound age estimate for Eurasian M and N is now 55 kya. Overlapping the age of major Eurasian Y chromosomal haplogroups C, F and D. And we have a proto-Eurasian genome from 45 kya whose Neanderthal admixture dates to this "recent OOA" period.

    Previously, the Skhul and Qafzeh AMHs were thought to have died out since the fossil record shows Neanderthals repopulating West Asia. Now, we find AMH admixture in Neanderthals that is unrelated to modern Eurasians (in fact, modeling San or Yoruba as the source works slightly better).

    I think Dienekes has to approach Dziebel levels of mental gymnastics to suggest these results point to anything other than recent OOA.

    btw, the last statement was a low blow! ;-) i hope you’re kidding.

    Read More
    ReplyAgree/Disagree/Etc. More... This Commenter This Thread Hide Thread Display All Comments
  20. @Razib Khan
    Does that tell us anything about where the ultimate cradle of modern human evolution was?

    i think only ancient DNA can resolve this. i have a broad belief that it was SS-african, but weak confidence of east vs. south (for example). and it could have been n. africa or near eastern. just lower probability IMO. though i'm pretty sure german dziebel is wrong.

    If it can resolve it at all. IIRC you were leaning towards a multiregional-in-Africa origin for humans, right? Do you still lean that way?

    Read More
    ReplyAgree/Disagree/Etc. More... This Commenter This Thread Hide Thread Display All Comments
  21. @Karl Zimmerman
    The problem here is the converse. We're now fairly sure there were archaic populations in Africa. We're also pretty sure that they interbred with at least some Sub-Saharan Africans not only after Africans had split off from non-Africans, but after those African populations themselves had been diverging for tens of thousands of years. Using the example of the pygmies from the other post, we are talking about AMH potentially living essentially alongside archaics for 170,00 years before a major admixture episode. There were not substantial geographic or climactic barriers across Sub-Saharan Africa over this entire time period. How could we have first diverged from the archaics, and then managed not to replace them (by whatever means) over such a long period.

    This is the crux of one of Dienekes arguments at least. If it wasn't for the basal haplogroup data being rooted so firmly in Africa, I'd find the argument that the foundation of AMH was somewhere else - somewhere with minimal influence by archaics - was plausible. But aside from A00, I don't find it likely any human haplogroups introgressed.

    On a broader note, these new papers have made me think about what exactly "AMH" means in discussing the human genotype. As an example, if you found a population like the Khulwilm paper - which split off genetically before the common ancestor of all modern groups, and left no living descendants - is it right to call it AMH? I mean, it is closer related to us than to Neandertals by descent, but Neandertals contributed to the modern human genome - this population did not. As archeogenetics gets to be more sophisticated, I expect we'll have more and more tough questions like this, since clearly applying the species concept, let alone such a fuzzy concept as "anatomical modernity" - to recent human evolution is by nature arbitrary.

    The problem here is the converse. We’re now fairly sure there were archaic populations in Africa. We’re also pretty sure that they interbred with at least some Sub-Saharan Africans not only after Africans had split off from non-Africans, but after those African populations themselves had been diverging for tens of thousands of years. Using the example of the pygmies from the other post, we are talking about AMH potentially living essentially alongside archaics for 170,00 years before a major admixture episode. There were not substantial geographic or climactic barriers across Sub-Saharan Africa over this entire time period. How could we have first diverged from the archaics, and then managed not to replace them (by whatever means) over such a long period.

    We probably did interbreed with them. We just don’t have ancient DNA to reference. If we were a set of constantly interbreeding populations we may be somewhat blind to this admixture I think? The Pigmy groups may have just encountered the most highly diverged/late persisting group so the signal sticks out strongest. If we never separated from some other groups in the first place… yah.

    Read More
    • Replies: @Razib Khan
    dienekes' contention http://dienekes.blogspot.com/2016/02/archaic-introgression-in-pygmies.html

    i do think that geo barriers which we think are trivial may not have been for earlier populations. the thousand years or so of farmer vs. HG 'apartheid' also points to how different human groups can avoid mating. perhaps with bigger differences separation could be maintained for longer?

    ReplyAgree/Disagree/Etc. More... This Commenter This Thread Hide Thread Display All Comments
  22. @CupOfCanada

    The problem here is the converse. We’re now fairly sure there were archaic populations in Africa. We’re also pretty sure that they interbred with at least some Sub-Saharan Africans not only after Africans had split off from non-Africans, but after those African populations themselves had been diverging for tens of thousands of years. Using the example of the pygmies from the other post, we are talking about AMH potentially living essentially alongside archaics for 170,00 years before a major admixture episode. There were not substantial geographic or climactic barriers across Sub-Saharan Africa over this entire time period. How could we have first diverged from the archaics, and then managed not to replace them (by whatever means) over such a long period.
     
    We probably did interbreed with them. We just don't have ancient DNA to reference. If we were a set of constantly interbreeding populations we may be somewhat blind to this admixture I think? The Pigmy groups may have just encountered the most highly diverged/late persisting group so the signal sticks out strongest. If we never separated from some other groups in the first place... yah.

    dienekes’ contention http://dienekes.blogspot.com/2016/02/archaic-introgression-in-pygmies.html

    i do think that geo barriers which we think are trivial may not have been for earlier populations. the thousand years or so of farmer vs. HG ‘apartheid’ also points to how different human groups can avoid mating. perhaps with bigger differences separation could be maintained for longer?

    Read More
    • Replies: @CupOfCanada
    Yah, I recall Dienekes' contention from the previous papers about archaic introgression in Pygmies, and other situations. I found his arguments persuasive on this, though I don't feel the same about an Out-of-Levant scenario.

    Fair point on those barriers. Still, you'd think there would have been at least some flow, no?
    ReplyAgree/Disagree/Etc. More... This Commenter This Thread Hide Thread Display All Comments
  23. @Razib Khan
    fwiw, the reason i dismiss lots of arcahic admixture as explaining african autosomal diversity is that you'd see a particular distribution of TMRCas of LOTS AN LOTS of haplotypes. imagine, for example, that 30% of khoisan ancestry is "archaic" and diverged ~500,000 years ago. 70% is "modern" and diverged 200,000 years ago. you'd have two clusters of gene phylogenies with different median coalescence dates to non-african humans, as the latter "modern" haplotypes segregating in the genome would merge more quickly to the last common ancestry than the large minority of "archaic" haplotypes. or, more visually, the haplotype trees would have two clusters across many genes. you don't see this to my knowledge.

    If populations weren’t well segregated, would it show this clustering? IE if there was a genetic cline of various types of humans across Africa, and the San accreted much of this cline as they moved south (assuming an origin outside of south Africa).

    Does the bulk of our DNA have TMRCAs of ~200,000 years? Is there anywhere to see a distribution of TMRCAs for various haplotypes across the genome as a whole?

    Read More
    ReplyAgree/Disagree/Etc. More... This Commenter This Thread Hide Thread Display All Comments
  24. @ohwilleke
    One of the biggest open questions in the genetic history of modern humans, in my view, is whether there is continuity between Skhul and Qafzeh AMHs or related populations that left Africa around then, or alternatively, if that was an "Out of Africa that failed."

    Keep in my the the datas you are referring to with regard to mtDNA M and N, and with Y-DNA C, F and D, is "the most recent common ancestor" date. Those most recent common ancestor dates can't rule out the possibility that people with the mtDNA clades of L3 that we now call M and N , and that the Y-DNA clades of Y-DNA B that we call C and F (and other sister clades of mtDNA L3 and Y-DNA B each that didn't survive a bottleneck) for many thousands of years before the TMRCA data.

    In general, lineages created by mutations are rarely lost when populations are expanding, are overwhelmingly lost and reduced during periods of declining population, and are pruned more swiftly than your intuition would expect during period of stable population. If you have an initial Out of Africa population that isn't huge to start with and declines between 125kya and 55kya or so, before suddenly expanding again, it is not very easy to distinguish this population based upon genetic data alone from one that leaves Africa in the first instance at 55 kya with mutation rate based dating.

    Linkage disequilibrium could easily distinguish the two scenarios, but this method doesn't have sufficient time depth to distinguish the scenarios in contemporary genomes, although LD methods applied to old Upper Paleolithic whole genomes might be able to make that kind of distinction.

    I personally favor a scenario with an initial Out of Africa ca. 125kya that survives in South Asian and Arabian interior populations for a prolonged period (possibly going extinct in Arabia), expands from South Asia to SE Asia and beyond ca. 75 kya following the Toba exclusion, and has a secondary expansion from South Asia (and possibly also Arabia) ca. 55kya in all directions (including SE Asia, Central Asia, Australia and back migration evidenced by mtDNA U6 and M1 in Africa) except for a Neanderthal barrier in Europe that persists until ca. 40kya with volcanic induced climate change gives AMHs a momentary edge that leads to Neanderthal extinction or near extinction.

    I think that the handful of older purported AMH fossils in China are probably either misdated or misclassified.

    Early OOA as the source of modern Eurasians does not work without substantial back-migration into Africa as well (not just Y-DNA CT but BT as well, not just mtDNA L3 but mtDNA L3’4, perhaps L2 as well), considering the more recent shared lineages. This is still possible, but it’s hardly the most parsimonious scenario.

    The recent paper dates mtDNA M and N to 47-55 kya, which should make L3 about 60-65 ky old. The branches from the L3 node are not that long.

    C, F, and D are somewhat longer branches from CF/DE (almost 20 ky removed per YFull). But if you look at the downstream subclades, C/F/D start bifurcating like crazy 48 ky, so interestingly pretty much simultaneously. This seems to correspond with the rapid Eurasian expansion, with the long branches perhaps being a result of the Eurasian bottleneck. Interestingly, African-affiliated E starts to split up a bit earlier than the other CT groups, around 54 kya.

    IMHO, this pattern would be consistent with an OOA around 55-60 kya, followed by a rapid expansion.

    Read More
    ReplyAgree/Disagree/Etc. More... This Commenter This Thread Hide Thread Display All Comments
  25. @Razib Khan
    dienekes' contention http://dienekes.blogspot.com/2016/02/archaic-introgression-in-pygmies.html

    i do think that geo barriers which we think are trivial may not have been for earlier populations. the thousand years or so of farmer vs. HG 'apartheid' also points to how different human groups can avoid mating. perhaps with bigger differences separation could be maintained for longer?

    Yah, I recall Dienekes’ contention from the previous papers about archaic introgression in Pygmies, and other situations. I found his arguments persuasive on this, though I don’t feel the same about an Out-of-Levant scenario.

    Fair point on those barriers. Still, you’d think there would have been at least some flow, no?

    Read More
    ReplyAgree/Disagree/Etc. More... This Commenter This Thread Hide Thread Display All Comments

Comments are closed.

Subscribe to All Razib Khan Comments via RSS