Detecting ancient admixture and estimating demographic parameters in multiple human populations

Detecting ancient admixture and estimating demographic parameters in multiple human populations (pdf)

A rather odd looking pdf- runs a lot like a mini PowerPoint presentation, by the man who published the recent paper that concluded there probably was archaic admixture in humans. And yet again.. I see the 40k-60k OOA date in print. Grr.

Detecting ancient admixture and estimating demographic parameters in multiple human populations

We analyze patterns of genetic variation in extant human polymorphism data from the NIEHS SNPs project to estimate human demographic parameters. We update our previous work by considering a larger data set (more genes and more populations), and by explicitly estimating the amount of putative admixture between modern humans and archaic human groups (e.g., Neandertals, Homo erectus, H. floresiensis). We find evidence for this ancient admixture in European, East Asian and West African samples, suggesting that admixture between diverged hominin groups may be a general feature of recent human evolution.

Yet another DNA study that finds evidence of archaic contributions in modern human groups. Odd how these don’t make the news but anything that finds in favour of the OOA gets splattered all over the media.

We estimate admixture proportions of 14 % (95% CI: 8 – 20 %) in the European-American sample and 1.5% (95% CI: 0.5 – 2.5 %) in the East Asian sample. In both cases, the relative log-likelihood for a = 0 (i.e., no ancient admixture) is significantly lower than the maximum likelihood (likelihood-ratio test, p < 10-3) , which provides additional evidence (along with the S* results in the previous paragraph) that ancient admixture occurred. The estimates of admixture rates in Europeans are consistent with estimates of Neandertal admixture obtained from analyses of Neandertal DNA (Serre et al. 2004; Noonan et al. 2006), [. . .] Unlike previous studies, we incorporated admixture between archaic and modern humans as an additional demographic parameter to be co-estimated. Interestingly, we could exclude no admixture (i.e., exclude a = 0) in both of the non-African populations studied

 The observation that all (three) populations studied seem to have evidence for ancient admixture suggests that ancient population structure may be a common feature of all contemporary human populations, and this ancient structure may predate the initial expansion of modern humans out of Africa.

Although some of the archaic DNA isn’t found in Africa, which would make the archaic admixture prior to the OOA hard to explain. This paper also finds evidence for archaic admixture in the Yoruba. I remember reading previously that the X chromosome showed signs of archaic ancestry in one pygmy group, so archaic ancestry in West Africa is supported by another paper. More detail… Testing for Archaic Hominin Admixture on the X Chromosome, which concluded the TMRCA was about 2 million years for one locus on the X chromosome and concludes..

For now, this locus represents a genealogical history that is most consistent with recent admixture from an archaic hominin population in Asia

Which is a far cry from Svante Paabo’s ‘no admixture but they were within the range of modern humans’ claim, which I found a bit odd. So you found they had essentially human DNA with us but decided they didn’t mingle …how?

I’d just like to comment that the OOA/RAR theory leaves absolutely NO room for any ancient DNA cropping in non Africans that doesn’t have a root in Africa- but it does, with remarkable frequency. In other words, the OOA doesn’t ‘fit’. That OOA is true of mostof our ancestry means sod all, it has to be true for all of it and it’s rather blatantly NOT the case, as there are a plethora of non-African but ancient in Eurasia mutations that invalidate it. Particularly the non African MC1R mutation ages that have ages of 100k-250k and a TMRCA of a million years.

Both African and non-African data suggest that the time to the most recent common ancestor is ª1 million years and that the age of the global 314 variant is 650,000 years. On this time scale, ages for the Eurasian distributed Val60Leu, Val92Met, and Arg163Gln variants are 250,000–100,000 years;

I’m going to have to make up a proper list of the DNA studies that find against the OOA theory.

15 responses to “Detecting ancient admixture and estimating demographic parameters in multiple human populations

  1. From what I know about Neanderthals they seem the least likely to have been the source of archaic admixture in modern humans.

  2. The first paper is just a modelling: there is no such “evidence” just speculative conclusions. As Hawks himself said, there are 9 interconnected different parameters which have to be guessed before the model can be run. The uncertainty is just too great, more as the populations used are just a minimal sample of all humans and, for example, early separated groups such as those leading to Pygmies or Khoisan are not even considered.

    I have read this study several times since it was in draft and has never sounded too convincing to me.

    The second paper was unknown to me but all they have is a TRMCA estimate, which I don’t see why would be reliable, much less for a single locus, which may perfectly be non-neutral (they have not the foggiest idea in fact).

    I you look at fig. 1 you can perfectly see how that TRMCA is plainly absurd: the youngest Eurasian only branch (H) “is” 300 ky old and the main mostly Eurasian branch (D) 600 ky old. Those figures are plain nonsense unless you go back to the multirregional model of different human species (not anymore just races) branching out from H. Erectus.

    At least they have a much better sampling strategy anyhow, with diverse South Asians being mapped as well as at least one East African sample: the Dinka. But I don’t see how can anyone match the absurdly long chronologies guesstimated here with anything that makes sense.

    A more logical conclusion would be to make their “1 MYA” = 100 kya (early OoA) and then the root would be c. 250 kya, in reasonable agreement with other known haploid data.

    The tree does not imply that A’B was in Asia before the A, B split, as it was just a private lineage whose endeavours are impossible to trace before that node (too easy to be lost in drift). Much like Y-DNA C’F for example (or if you press me, like both C and F separately in the long coalescent period before they branched out, already in Asia).

    If my “half of the half of the half” or more like 10% of what they claim as ages is correct, then the age of H would be just 30 kya and the age of D would be like 60 kya.

  3. “That OOA is true of mostof our ancestry means sod all, it has to be true for all of it and it’s rather blatantly NOT the case”.

    I find it absolutely amazing that so many people are so absolutely opposed to the idea of admixture. I can understand creationists opposition, but scientists’?

  4. Were Neanderthal that similar to homo sapiens ?
    Have they been any incidence of primate cross-breeding? Or is it a case of nothing else around and close enough situation!! lol

  5. Mathilda,

    Speaking of your “grrr” reaction to the 40,000 years ago Out of Africa date, I just came upon an entry of mine from December 2007 that you might find amusing:

    The ever-shifting date of the African exodus

    A few months ago I read Michael Hart’s Understanding Human History (reviewed by me here), which said that a small group of modern humans—the ancestors of all non-African humanity—left Africa 60,000 years ago. While I had previously understood that the date of departure was considered by scientists to be 100,000 years ago, I internalized Hart’s new figure of 60,000 years ago and proceeded to use it in my own thoughts and discussions. Then I read Nicholas Wade’s Before the Dawn, which says that the proto group of modern humans left Africa 50,000 years ago. I was a little thrown that there would be such a variance on an event about which the evolutionists claim to be so certain, but I adjusted and began using that figure. Now Steve Sailer at Vdare quotes Henry Harpending to the effect that modern humans left Africa 40,000 years ago. Please, guys, can’t you get together on this?

    By the way, Wade says that modern Homo sapiens did not differentiate into distinct races until 12,000 years ago. This seemed like the most sensational and unlikely claim in his book, but I haven’t seen any discussion of it.

    Posted by Lawrence Auster at December 31, 2007 09:04 AM

  6. Mathilda you are wading through a LOT of material which is appreciated, and a more-or-less complete list of non-OOA DNA is eagerly anticipated.

    I wish I had the time to do some of this… I would especially like to see all of this compiled so that we could see just what we mean by “ancient” DNA…

    I am forced by the lack of mention of H. heidelbergensis to become an advocate for this species. Neandertal, neandertal, neandertal!!! If earliest Homo sapiens did indeed arise in a single location (so far we have from 200,000 bp only H.s. idaltu from Ethiopia, right?) and if the modern post-L1 likewise, and also from Ethiopia, then as they spread around Africa, would they not have mixed with the extant humans, which were H. heidelbergensis?

    Rather than ramble on, here I list the possibilities for admixture of which I am aware.
    1. earliest H. sapiens with H. heidelbergensis in Africa
    2. earliest H.sapiens in SW Asia with H. heidelbergensis
    3. earliest H.sapiens in SW Asia with H.neandertalensis
    4. H. sapiens in Europe with H. neandertalensis
    5. First wave of H. sapiens in India & E. Asia with H. heidelbergensis
    6. First wave of H. sapiens in E. Asia with H. erectus
    7. Post-L1 H. sapiens in Africa with Lo and L1 H. sapiens
    8. “OOA” post-L1 H. sapiens (70K bp and later) in Eurasia with surviving earlier wave(s) of H. sapiens (not all wiped out by volcano 72K bp)

    I will continue to b—- and moan about the use of words like “archaic” and “ancient” during these discussions. The widespread use of the term “archaic sapiens” caused plenty of confusion, so that nowadays even in the utterances of anthropologists there is much confusion about whether they are referring to sapiens or heidelbergensis. I could make another list of how the various non-taxonomic terminology confuses the issue.

    I just saw Wolpoff re-run again on the History channel saying “neandertals are us”… Well, folks, let’s not throw out the baby with the bath. Because 100% of modern H. sapiens DNA is not of OOA origin (I think the various DNA studies cited by Mathilda here are enough to at least prevent anyone from ASSUMING 100% OOA) does not mean that we are not essentially OOA humans.

    My opinion (at least this week) is that certain alleles entered the OOA (post-L1) population from prior Eurasian and African Homo spp and were selected for. On the whole, pre-OOA DNA occurs in small %. As with the proliferation of the Natufian wheat-tolerant gene in the Mideast, where little other DNA evidence of Natufians remains, the advantageous genes were perpetuated, but the total relative number of pre-OOA ancestors was never large.

    Not statistically significant, perhaps, but just as the Natufian wheat gene enabled significant changes in human culture, the non-OOA DNA has likely proven to be very important in the physical and cultural evolution of both African and OOA humans.

    I’d be interested to know how many on both sides of the OOA argument could accept the above paragraph. Science like politics is often impeded by entrenched positions based on varying interpretations of data… The truth has to involve some melding of the 2 positions, as the extremes on either side are untenable.

    • I’ll get onto all it at some point. I’ve not been well lately so it’s all having to go on the back burner for now.

      Mainly OOA people.. and the date is about 100k…

  7. I strongly second Sinajuavi’s complaint about the confusions created by the current use of “archaic.” Here is an e-mail I wrote about this yesterday to Ronald Fonda, whose articles I’ve been reading:

    Consider this incoherent paragraph from Wikipedia. I always thought that archaic Homo sapiens was a sub species of Homo sapiens, just as we are. But, according to this, by “archaic Homo sapiens” they mean NON-SAPIENS SPECIES. So now everything is thrown into confusion. How can Homo neanderthalenssis be archaic Homo sapiens??? What happened to the archaic form of H. sapiens who I thought lived from about 200 kya to 100 kya, and whose skull was pretty much like ours, except a bit more robust, and with a small brow ridge? In this account, he’s dropped out of the picture, and archaic H. sapiens now refers to neanderthal, rhodesiensis, etc. Ridiculous.

    Archaic “Homo sapiens” is a loosely defined term used to describe a number of large hominids that emerged by about 500,000 years ago. Archaic Homo sapiens typically refers to Homo heidelbergensis, Homo rhodesiensis, Homo neanderthalensis and sometimes Homo antecessor.[1] Modern humans are believed to have evolved from archaic Homo sapiens, who in turn evolved from Homo erectus. Archaic homo sapiens are referred to as “Homo sapiens” because their brain size is very similar to that of modern humans. Archaic Homo sapiens had a brain size averaging 1200 to 1400 cubic centimeters, which overlaps with the range of modern humans. Archaics are distinguished from anatomically modern humans by having a thick skull, prominent brow ridges and the lack of a prominent chin.[1][2]

    On the same Wiki page, a photo with this caption:

    “Anatomical comparison of the Skulls of anatomically modern humans and homo neanderthalensis”

    Now this is absurd. The very meaning of the phrase “anatomically modern human” is to distinguish it from archaic Homo sapiens. If you’re distinguishing H. sapiens from neanderthal, then obviuosly the distinction should be expressed as:

    “Anatomical comparison of the Skulls of Homo sapiens and homo neanderthalensis”

    When an entire field starts using words with such shaky and contradictory definitions, that is a strong sign that the people in that field are not thinking clearly.

  8. @Terry: the data is the data, man. I think people who insist against all evidence for hybridation are the ones who are being religious. Judaism is not the only possible religious line of thought, Neanderthalism can be another. For science only the data matters and the objective data is so far strongly against significative hybridation (insignificant one cannot ever be discarded).

    @Lawrence: c. 50 kya would be the most recent possible date, not for the OOA strictu sensu but for the subsequent expansion in Asia. I think the fossil record is conclusive enough to not allow anything more recent than that. When someone says “40kya”, he/she is breaking with the most recentist possible assumption, because even European colonization is older than that. If you ever read that again, you can stop reading at that point: whoever writes that is a total ignorant.

  9. “The ever-shifting date of the African exodus”.

    Strongly suggests that the African exodus was no simple matter. However we all like to find simple answers to complex questions.

    Mathilda. Sorry to hear you’ve not been well lately.

  10. Warm Evolution

    – “Evolution faster when it’s warmer” –

    Could explain why there is so much more genetic diversity amongst equatorial Africans as opposed to the White Europeans and Asians who live in colder northern climates.

  11. Are you alright, Mathilda? One whole month without a single post and last thing I know was that you were complaining from pains.

    Hopefully you have just taken a well deserved vacation but concerned in any case.

    • My MS was playing up- blurred vision, fuzzy head, left hand’s partially numb agian. Blogging was not on the agenda as I couldn’t focus on the screen. I’m just wading through my back comments.

  12. Apologize my English please.

    I have questions.

    1. According to genetics, the biggest changes mutations mtDNA are in black race. So, black race is the oldest race in the world. But, if the other races are from black race these mutations must have too. And if the have not we cannot are from black race.

    2. According to genetics research race are not from Cro Magnon man, who appear in 35 000 – 30 000 before Christos. This is very short time for creation of races. Besides we have not found any development forms of him. But a part of men is from Cro Magnon man.

    So, tell me someone how could we come from black race???

  13. Sinajuavi, I agree with your list of possible admixture events. And I think most people would agree that mtDNA L3 spread through an Africa already largely occupied by humans with mtDNAs L0, L1, L5 and L2, basically your number 7. But mtDNA L3 may have been the first modern haplogroup able to expand across what had become a relatively moist Sahara, and so emerge from Africa.

    I’m also inclined to suspect that the bulk of us humans descend from a hybrid formed somewhere after the first OoA exit, probably round the Zagros Mountains, or on the Iranian Plateau, basically your numbers 2, 3 and 5. This hybrid then spread back into Africa, into India and around the world, forming further admixtures with pre-existing populations, the remainder of you admixtures. This would explain the apparent rapid diversification of phenotype. The resulting complex pattern does not enable us to separate OoA genes from those already widespread, although it’s possible some people at the southern tip of Africa lack any genes from outside that continent. And so to isolate original OoA genes perhaps we should compare the genetic makeup of Khoi-San with that of people far removed from Africa, perhaps Melanesians or Native Americans.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s