Substack cometh, and lo it is good. (Pricing)

The details of Eurasian back-migration into Africa

Carl Zimmer has an interesting write-up on the new method to detect Neanderthal ancestry in Africa, Neanderthal Genes Hint at Much Earlier Human Migration From Africa. There are two quotes from researchers that are of note.

First, from David Reich:

Despite his hesitation over the analysis of African DNA, Dr. Reich said the new findings do make a strong case that modern humans departed Africa much earlier than thought.

“I was on the fence about that, but this paper makes me think it’s right,” he said.

It’s possible that humans and Neanderthals interbred at other times, and not just 200,000 years ago and again 60,000 years ago. But Dr. Akey said that these two migrations accounted for the vast majority of mixed DNA in the genomes of living humans and Neanderthal fossils.

Over the years I have had several discussions with members of the Reich lab about whether there was a major migration of the antecedent lineage of modern humans before the one that we detect 60,000 years ago. Many were quite skeptical because of the lack of clear genetic signal of anything before 60,000 years ago, as well as its correlation with a strong archaeological record. But, it seems now that David Reich at least is convinced that the evidence of admixture into Neanderthals means that there were descendants of the same lineage that led to the major “Out of Africa” expansion 60,000 years ago who had spread earlier (though the footprint was small, and their impact on later humans difficult to detect).

Second, Sarah Tishkoff says something that I forgot to mention in my earlier post:

Sarah Tishkoff, a geneticist at the University of Pennsylvania, is doing just that, using the new methods to look for Neanderthal DNA in more Africans to test Dr. Akey’s hypothesis.

Still, she wonders how Neanderthal DNA could have spread between populations scattered across the entire continent.

The second part isn’t that inexplicable. In the paper, they mention that they don’t have the power to analyze small sample numbers. So they focused on the 1000 Genomes samples, which are from West and East Africa. From agriculturalist and agro-pastoralist populations. If you listen to this week’s episode of The Insight Spencer and I talk extensively about the recent agriculturally mediated expansions within Africa. Much of the genetic landscape of the continent is novel, new, and of short historical time-depth. The Africa of Old Kingdom Egypt, 4,500 years ago, was very different.

As hinted by Tishkoff the key is going to be when we get samples from hunter-gatherers. Some of these have much lower Eurasian affinities, and likely they’ll carry less Neanderthal ancestry.

On a final note, this paper and the first author, Joshua Akey, hints at some resolution in the interminable disagreement about continuous gene flow vs. pulse admixture. Some of the methods to infer and detect admixture assume pulse admixture, and so our conception of the past has been skewed. On the other hand, I think it is plausible that in a patchy low population density Paleolithic landscape continuous gene flow may have been quite attenuated over long distances. Admixture then would occur when there were cultural revolutions and long-distance contact for short periods of time, before an equilibration. Basically, it’s some of both.

22 thoughts on “The details of Eurasian back-migration into Africa

  1. The main impression from the comments of the scientists I get is the reluctance to accept what was for quite some time, in my opinion always, the most likely scenario: That there was no simple, fairly recent out of Africa to begin with and that there were massive migrations out and into Africa.

    It reminds me on the debate about “pre-Clovis”. Its not like it wasn’t known before that there are older human remains and different layers cultures and people in America. The evidence was sparse, but there was no reason at all to believe there was just one major immigration and nothing before.

    This was made a dogma at some point, for a variety of reasons, including ideological ones, and then it was kept alive when the evidence against was mounting.

    However, science and scientific ethos are at least alive, even if they are not doing well, as long as people work hard to get to the truth, accept it and correct their own views. Honest people is what we need. Respect for those which change their opinion when new facts appear and stay to their new conclusions even if its less popular than sticking to the old paradigm.

    “No Pre-Clovis” and “strict Out of Africa” are practically dead by now and its for good. I had my opinion about archaic introgression before genetic testing was available. I refuted it completely, didn’t believe Wolpoff & Co. Not that he was completely right in every respect, but more right than most. He was on something and sticked to it, even though he was in the minority for quite some time.

    Only new data could make the difference. Now we need some more samples from Africa to complete the puzzle and get the full picture for recent human evolution and gene flow.

  2. “As hinted by Tishkoff the key is going to be when we get samples from hunter-gatherers. Some of these have much lower Eurasian affinities, and likely they’ll carry less Neanderthal ancestry.”

    Aren’t Neanderthal affinities found at similar levels whether we’re looking at (non-Fulani) West Africans, Pygmies or Khoisan?

  3. @Lank: That would be against the trend, because Pygmies and Khoisan have significantly more from an earlier Homo sapiens branch (Ghost modern in Shum Laka paper plus archaic). So the difference might be not that big at all, especially between Pygmies and and West Africans, but it should be noticeable.
    In the case of Khoisan we have to consider that especially Khoikhoi have significant more recent Negroid and Caucasoid admixture and San too, even if on a very low level.

    Its just a wild guess, but if the current models are correct, Pygmies should have about 50-60 percent Neandertal in comparison to West Africans (Niger-Congo/Yoruba) and San probably about 20 percent. So not no, but significantly less.

    Remind you that Shum Laka itself was mixed already and modern Pygmies too have significant Bantu and Niloto-Hamitic ancestry. So they are no representatives of the “pure African foragers” of West and Central Africa before the back migration began. Those don’t exist any more and are a true ghost population like “Basal Eurasian” for West Eurasia. They might have ceased to exist in a pure form even long before the Shum Laka individuals.

    But if one population should show the lowest level of Neandertal ancestry, it should be definitely the San, because of all people, they branched off earlier and have (not no afaik!) but the lowest amount of admixture from proven Neandertal admixed people.

    So if the level of Neandertal ancestry would be the same in all SSA, there would be something missed or completely wrong about the method. I doubt they didn’t check.

  4. Unfortunately, this paper only looks at Niger-Congo speaking Africans. The level of Neanderthal detected was 1/3 of that found in Eurasians, which would be quite significant. If modern San only have 20% of that (6-7%), then that’s an important difference and should be detectable.

    This needs to be looked into further and with more samples before drawing any premature conclusions. Interesting methodology anyhow. I will stay cautious, as David Reich also seems to be, based on the NYT article:

    David Reich, a geneticist at Harvard Medical School, praised much of the study but said he had doubts about how extensive the flow of DNA back to Africa could have been. “It looks like this is a really weak signal,” he said of the data.

  5. @Walter, if I’ve got it right, looks like either the NeanderSovans* entered Eurasia from Africa at some undefined point after splitting with modern human lineage, or the modern human lineage entered Africa from Eurasia at some undefined point after splitting from NeanderSovans (and then migrated back into Eurasia at some point, possibly in waves some of which mattered and others of which didn’t, probably big expansions less than 100kya ago).

    It seems impossible to tell between these two scenarios, without much older adna from Eurasian and African hominids than we have and which I think that we are unlikely to ever get in any form, let alone the quantity needed.

    According to “Who We Are And How We Got Here”, NeanderSovan split with AMH probably something close to 660kya (against intra-AMH split estimate something like 270 kya). Though there are other estimates.

    Admix from Neanderthal is specifically from Croatian Vindija Neanderthal over Altai Neanderthal (and Neanderthal over Denisovan) so apparently can’t really be explained well by population structure in AMH ancestors in Africa.

    *common ancestors of Neanderthals and Denisovans, who split but largely form a clade to humanity, barring these low level admixture events that we know about into AMH populations, including some admix from AMH ancestor into Neanderthal and not Denisovan, and some small “super-archaic” admixture into Denisovans.

  6. @Walter: The basic assumption is still that Homo, especially Homo erectus, evolved in Africa and spread from there. The different branches leading to Homo neanderthalensis (Europe/West Asia), mixed/evolved Homo erectus/denisovan (North, East and SEA) and Homo sapiens (North and East Africa?) evolved from earlier Homo erectus strata. Exact migrations, replacements and exchanges largely unknown. Some split H. e. in more groups, but I don’t think that makes a big difference at all – there are of course intermediate forms, like Homo heidelbergensis, but that too doesn’t change the bigger picture.
    But as things are, who knows for sure? Some months ago nobody knew about Danuvius guggenmosi and there might be a lot more surprises to find.

    @Lank: For the model to be reliable, San need to have less Neandertal than Yoruba. That’s conditional. I can’t imagine they didn’t check, that’s unthinkable to me, especially after the involved authors said they double checked everything before trusting their own results.

    And by now it is very clear, also from the Shum Laka paper, an mtDNA paper from 2018 (E + L3 back-migration from Eurasia) and now this Neandertal introgression model, that the Eurasian or at least Eurasian related replacements in Africa didn’t affect all African populations the same way. It should decreases like that:
    North Africa -> East Africa -> West Africa -> Central Africa -> Pygmies -> San.
    The Neandertal admixture traceable should just reflect that pattern. Compare with these comments:
    https://www.gnxp.com/WordPress/2020/01/30/neanderthal-ancestry-in-africa-and-west-eurasian-gene-flow/#comments

  7. The above said, I think there is an argument that the radiation of the ancestor of AMH+Neandersovan should have happened from Africa because there are some fossils at/near the estimated split time that show a higher cranial capacity in Africa (Bodo).

    But I am not sure how well this actually holds up when we look at the general trends over time and sparse fossil records.

  8. Sparse fossil records is the issue, if we missed prehuman bipedal primates from Eurasia altogether until recently.
    Just one decisive find can turn everything upside down. To talk about gaps is an euphemism, considering we just detected small islands in a sea of ignorance so far. Later humans seem to have been quite numerous and used sheltered positions. But that was not the case for earlier relatives.

  9. @Obs: There’s no mystery here, they used samples from the 1000 Genomes Project, which has samples from West Africa and Kenya, not Pygmies or Khoisan.

    And it’s far from clear that Y-DNA E, let alone mtDNA L3 are from Eurasia. You have rather lax standards of evidence when it comes to Eurasian admixture in Africa.

  10. @Obs: The hypothesis of this new paper (Chen et al 2020) is (in agreement with much other research) that, despite some significantly earlier migrations from Africa of H. sapiens/AMH (earlier OOAs), the (majority of) ancestors of modern Eurasians left Africa ca. 60kya and then admixed with neanderthals. And then (Chen 2020 suggests) early West Eurasians back-migrated to Africa later (ca. 20ky), after the split between Western Eurasians and Eastern Eurasians and admixed with Africans. This is quite different/distinct from the Eurasian L3 hypothesis of the 2018 paper (Cabrera et al.) you referenced, which proposes that Eurasians descend mostly from a 125kya OOA migration that then back-migrated to Africa around 70kya (which most research does not seem to support), which then carried haplogroups L3, DE, and E to Africa (as part of that putative ca. 70kya back-migration). However, most evidence indicates that there is little to no pre 60-50kya OOA wave ancestry in most modern Eurasians (those earlier OOA waves likely having died out or been absorbed by Neanderthals, with only about 2-3% remaining in Oceanian peoples like Papuans and Australian Aboriginals).

    A 2019 study (by Haber et al., link below) calculates divergence dates for haplogroups DE and E (and newly-discovered DE branch D0), both within about 76-71kya) that predate the migration of the majority of the ancestors of modern Eurasians from Africa which it also estimates at about 60-50kya. The study concludes that y haplogroups E, DE and D0 are likely to have diverged somewhere within Africa (though one might guess somewhere in east or northest Africa) before the (major) ca. 60-50kya OOA from which most Eurasian ancestry derives.
    https://www.genetics.org/content/212/4/1421

    Regarding, L3, the 2018 Cabrera paper you cited is partly based on the idea that the most basal branches of mtdna M and N are situated in/near southest Asia far from Africa, and that their immediate ancestor (L3) thus could not likley have come from Africa. A more recent study finds a basal branch of N in the early neolithic/late mesolithic Libyan Sahara (from ca 7kya), and suggests that L3 diverged in East Africa, with M and N diverging soon after in the nearby Near East (the paper suggests Arabia), also suggesting a North African origin for N as an alternate possibility. See below:

    “Ancestral mitochondrial N lineage from the Neolithic ‘green’ Sahara:
    https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6401177/

    The population from which E, DE, and L3 came may not have been Eurasian, but it very well could likely have been an African population (likely from east or northeast Africa) related to the African ancestors of Eurasians (not necessarily Eurasian, but perhaps, as you suggested, Eurasian-related; possibly from a population similar/related to Mota or to some now-extinct indigenous north/northest African population that diverged soon before the major OOA).
    The 2020 Shum Laka paper estimates a ca 80-70kya divergence (in Africa) for the common ancestors of Eurasians and most non-Khoisan/non-Pygmy Africans (which includes Niger-Congo and Nilo-Saharan peoples). This would likely be the ancestral population of Eurasians/non-Africans, much of the ancestry of East Africans, and the non-Basal ancestry in West Africans.

    @Lank
    It would also be interesting to see analyses of Nilotic (and other Nilo-Saharan) groups, e.g. the Dinka, who overall cluster somewhat closely with Niger-Congo populations but generally lack ydna E at significant levels (whose Eurasian origin is, as you have said, far from clear, but has been suggested).

  11. Minor Edit to previous comment: “Regarding, L3, the 2018 paper…is partly based on the idea that the most basal branches of mtdna M and N are situated in/near southest Asia far from Africa, that their origin is likely to be in/near southest Asia, and that their immediate ancestor (L3) thus could not likley have come from Africa. A more recent study finds…”

  12. @Jm8: I know the paper includes some controversial ideas, not just what you wrote, but also the Northern expansion route and a Central Asian origin of the basal split etc.

    However, what this paper is really showing and the more recent studies, including the mtDNA study on Africans I quoted before support, is that haplogroup E and L3 are not from West Africa, but came in with an Eurasian related ancestral component which expanded MASSIVELY into SSA.

    Now this is the baseline for any debate on the issue. With the new data we see that at least one major element of this Eurasian-related ancestry was Eurasian proper, because it brought into Africa Neandertal introgression. Looking at the candidates of all these papers, you have not just, but primarily, yDNA E and mtDNA L3 as candidates. I don’t know in detail how this will pan out, that’s why I was careful and spoke about one possible scenario of E picking L3 in North East Africa up and taking it to the West.

    But as things are, the calculations for E and L3 are not fully reliable, but if the IBDmix estimate is correct, we have a need for at least 30-40 percent backflow from Eurasia. So that there was actual backflow, not just North East African or East African, Eurasian related dispersions, is, going after that result, a given fact. There are just these two options: IBDmix is right, than you need to account for this, or they are wrong.

    L3 can be questioned, but yDNA E not. I think that’s well established now. And L3 will become too, because while I trust the importance of sex biased admixture, it seems to be more parsimonious to assume a maternal companion. They could have picked it up in the North East and a mixed E : L3/L2 population could have first populated the green Sahara, then moving South under pressure from the E : U6 more West Eurasian Northern group, which we see in Taforalt. I won’t exclude that now, only new samples can fix that.

    The paper you linked: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6401177/
    Just supports the idea of an expansion through the Green Sahara and makes just some of the claims from the 2018 paper questionable – probably. But its still in full agreement with a proposed E : L3 expansion from Eurasia or alternatively an Eurasian/North East African borderzone.

    Its very, very unfortunate they sampled not for autosomal or yDNA! Because this is exactly what we might have needed, searching for haplogroup E, especially now SSA E-M2. From the paper:

    “A possible scenario envisages an introgression from Eurasia in ancient times that carried haplotypes that have since disappeared from Africa. The timing of this migration remains difficult to define. Late Pleistocene dispersal from Western Asia into Africa around 39–52 ka is suggested by the expansion of the U6 haplogroup29,30, with a potentially corresponding archaeological signature in the MSA Dabban industry of Cyrenaica, Libya, ca. 45–40 ka31. Individuals carrying a N haplogroup basal lineage could have followed the same dispersion pattern as U6: their legacy could have been survived up to ∼7000 years ago in the central Sahara thanks to the climatic conditions previously described, but replaced and disappeared in other parts of North Africa.”

    https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6401177/

    They also propose a relationship to Subsaharans morphologically. So this could be at least a relative.

    My guess: Complex mixture processes in the Green Sahara region, a push from Eurasian proper from the North and worsening climatic conditions led to the only route the Sahran pastoralists could take: South. Niger-Congo/Bantu is the result of this Saharan expansion South.

    The question is just how much was coming directly from Eurasia and the Neandertal admixture, if proven, would say a lot.

  13. It would also be interesting to see analyses of Nilotic (and other Nilo-Saharan) groups, e.g. the Dinka, who overall cluster somewhat closely with Niger-Congo populations but generally lack ydna E at significant levels (whose Eurasian origin is, as you have said, far from clear, but has been suggested).

    in personal communication over lunch david reich told me he was perplexed by how little eurasian ancestry dinka seem to have. it seems clear that ‘eurasian ancestry’ isn’t a function of distance from eurasia. the hausa for example clearly have more than the dinka.

  14. @Obs:
    You wrote: “but if the IBDmix estimate is correct, we have a need for at least 30-40 percent backflow from Eurasia”

    Chen et al estimate about 0.3% Neanderthal admixture in Africans, while Eurasians have on average about 1.8-2.5%, or 2%. Since .3 is about 15% or 1/6th of 2, this would seem to suggest about/roughly 15% Eurasian admixture in Africans (give or take).

    But I would agree with Lank that it is too early to be certain regarding the study’s conclusions (as David Reich remarks); they could turn out to be correct or not (or it may be that Neanderthal/Eurasian admixture could be present, but not necessarily at the levels proposed by Chen et al.).

  15. @Razib Khan: Yes, that is interesting (though not too surprising since the Hausa are Afro-Asiatic-speaking and fairly rich/high in the Eurasian-derived y-haplogroup R1b, as well as having, I believe, some E1b1b). It would be interesting to see Dinka/un-mixed Nilote results using Chen et al.’s method (as used in this recent paper).

  16. @Jm8: At page 8 they write:
    “Accordingly, African individuals have approximately 33% as much detected sequence compared to non-African individuals.”

    The about one third Neandertal ancestry in West Africans being repeated in the tables of the supplement.

    So is the Neandertal ancestry going after the study 1/3 of WEA/EA in Yoruba or not?

    Also:

    “Furthermore, it is increasingly recognized that geneflow occurred among structured populations across the African continent, Eurasian ancestry is found across Africa (Pick-rell et al., 2014). Even early diverging groups like the Khoe-Sanhave up to 30% ancestry from recent admixture with East Afri-cans and Eurasians (Schlebusch et al., 2017). Therefore, it willnot be surprising if Neanderthal ancestry, due to back-migra-tions, is present at varying levels across the African continent.”

    That Nilotes are supposed to have such a low Eurasian admixture is surprising indeed and needs further inspection. Also which other admixtures West, Central and South Africans might have, but Nilotes not.

  17. @Obs
    You wrote:
    “But its still in ” …[the 2019 Vai et al. paper on basal N in the Sahara]… in full agreement with a proposed E : L3 expansion from Eurasia or alternatively an Eurasian/North East African borderzone.”

    Possibly with the latter, but that paper does not posit a Eurasian origin for E or L3. It does not (or does not much) discuss Ydna haplogroups, but does suggest/propose an origin for mtdna L3 in Africa (Figure 4, in Vai et al. 2019), though suggesting a most likely origin in Eurasia/West Asia for N and M (L3’s daughter haplogroups) after the (ca. 65-55kya) OOA-associated migration of L3 to West Asia, with N likely later back-migrating to north Africa (or alternately originating/diverging from L3 there – though the paper seems to favor the first scenario: a West Asian/Near Eastern divergence of N followed by its back-migration to North Africa).

  18. @Jm: Right, but the paper does nothing to refute the idea of L3 coming back from Asia. It just proves the presence of N.
    This is somewhat at odds with the other paper, but not at all a biggy. In reality we need more samples until this debate can be resolved in a satisfying way. Anything else will remain speculative.
    But an origin of L3 in Eurasia remains a possibility, we are just debating probabilities. Nothing definitive.

  19. Regarding African mtDNA N and y-DNA D0 – the Libyan N is still downstream of Oase’s N, so I don’t think just because it’s basal to modern N lineages means N itself formed in Africa. And regarding D0, if you take its presence in West Africa as a point of evidence in favor of DE originating in Africa, then you have to explain why one branch of DE (y-E) totally replaced its sibling clade in virtually all of Africa save for a handful of stragglers. F derivatives are by the far the most dominant Y clades in Eurasia, but there’s still plenty of C and even D still around. In Africa E is dominant but there’s still respectable levels of A and B. If DE emerged in Africa, why was D virtually wiped out by E, but not A or B?

  20. Hi, Razib It‘s a bit off-topic but I‘d love to hear your opinion about the Shum Laka Ancient DNA paper that appears to shake certain ideas about Eurasians and the SSA population.

    There are people who claim that the Shum Laka study shows that ANA is on the same branch as Basal West African and that the most critical conclusion is that the component that constituent the largest component of West Africans (Basal West African) is on the same level and radiated with Eurasian. So basically West Africans are in the same branch with Eurasians with Khoisan as an outgroup. These people argue that the “exclusivity” of the so-called Eurasians in Africa, got a heavy blow with the Shuma study since any “Eurasians” who back migrated back to Africa wouldn‘t have been “Special”. They just would have met African people who were not that different from them, people who later picked up other ancestries with time. Hence the terms like Subsaharan African and Eurasians are totally meaningless genetically, particularly in Africa.

  21. “it seems clear that ‘eurasian ancestry’ isn’t a function of distance from eurasia. the hausa for example clearly have more than the dinka. ”

    Some Hausa tribes/ people have absorbed Fulbe/Fulani. Plus there are also Hausa people that are basically assimilated Fulanis or strongly admixed with them. So it’s very likely that most Eurasian admixture in Hausa is Fulbe/Fulani mediated. Hence the Hausa harbor higher Eurasian mix compared to the Dinka. I assume that there are Hausa people that have the same low level of Eurasian as the Dinka tribe. Just saying.

Comments are closed.