September 06, 2011

Archaic admixture in Africa confirmed

I have been somewhat of a broken record in arguing for archaic admixture in Africa over the last few years. My reasons for doing so have been stated many times, but here they are, once again:
  1. African populations are structured today, why would they be otherwise at the time of the emergence of Homo sapiens?
  2. Africa has a great deal of ecological variability, so the emergence and spread of H. sapiens from some region of the continent is consistent with the presence of other H. species elsewhere
  3. Africa has an extremely long hominin record, and a diverse climate. It is inconceivable that archaic H. forms there would simply vanish, as they were well-adapted in their particular niches
  4. African populations of today show the greatest genetic diversity: but, to what extent is that a consequence of them being the ur-humans, and to what extent is it the product of admixture between distantly related H. populations in a multiregional model?
  5. The palaeoanthropological record has important lacunae in much of tropical Africa where conditions for preservation are extremely unfavorable; so, it is possible that there are as of yet undetected H. populations that lived there
  6. Even the fossils we do have do not paint a picture consistent with the naive Out-of-Africa model. For example, H. sapiens idaltu (Herto) was hailed as our ~150ky ancestor when it was published, only to lose that title by the redating of Omo to ~195ky. Omo is more modern than Herto anatomically, so this is consistent with the idea of modern humans not being the "only game in town", when they make their entrance
  7. Indeed, unquestionably modern, but archaic-looking forms persist in the available record down to a few thousand years ago, such as the Fish Hoek and Boskop samples
All of the above, however, amount to nothing, without a real genetic smoking gun. In Africa, because of climate, we are unlikely to ever extract DNA from really ancient hominins, as it was possible in Eurasia with Neandertals and Denisovans. Moreover, in tropical regions, such as the jungles of Central Africa, the odds are stacked against finding (m)any H. specimens. So, the evidence has to come from modern genomes.

Hence, I was pleasantly surprised to see a new paper appear in PNAS whose title pretty much summarizes what I've written about for so long. At present, the paper is not "live" on the PNAS site, so I will refrain from blogging about its actual contents. There is already a story in Nature:
Hammer and his colleagues argue that roughly 2% of the genetic material found in these modern African populations was inserted into the human genome some 35,000 years ago. They say these sequences must have come from a now-extinct member of the Homo genus that broke away from the modern human lineage around 700,000 years ago.
and a press release:
Hammer said that even though the archaic DNA sequences account for only two or three percent of what is found in modern humans, that doesn't mean the interbreeding wasn't more extensive.

"It could be that this represents what's left of a more extensive archaic genetic content today. Many of the sequences we looked for would be expected to be lost over time. Unless they provide a distinct evolutionary advantage, there is nothing keeping them in the population and they drift out."

In a next step, Hammer's team wants to look for ancient DNA regions that conferred some selective advantage to the anatomically modern humans once they acquired them.

"We think there were probably thousands of interbreeding events," Hammer said. "It happened relatively extensively and regularly."
We must note that detection of archaic admixture is a bit like detective work. For ancient DNA, it involves dealing with contamination and damaged DNA; for inferring admixture from modern DNA it involves statistical modeling and is inhibited by drift (which removes all evidence of admixture) and recombination (which breaks up archaic segments that introgressed long ago into small chunks in present-day humans).

These factors are often not emphasized enough:
  • Small chunks of DNA are more difficult to assign to a modern or archaic origin; a long stretch of DNA with multiple substitutions relative to most modern humans can be proven not to have arisen by simple mutation in a conventional OoA timeframe, but this becomes increasingly difficult the smaller the segment
  • Recombination breaks up DNA into smaller chunks with the passage of time; given enough time there are no longer any identifiable archaic segments but only inconspicuous remnants of archaic DNA that no longer "stand out" as peculiar against the background
  • Drift gets rid of modern and archaic DNA alike. A million years from now most of our DNA will be extinct because of it. Archaic admixture that happened a long time ago can no longer be detected.
Just like a detective doesn't catch most criminals but makes sure that the one he catches are criminals, so does the genetic detective try to catch real archaic DNA, but, in the process misses a lot of archaic DNA if there is no evidence for it, or it has gone missing by drift and recombination.

The conclusion is obvious: archaic admixture is adding up: a few percent in Africa, a few in Eurasia, a few in Australasia. All this is the tip of the iceberg. With the next ancient hominin to be unearthed, with the next collection of full human genomes, with the next suite of more powerful statistical tools, archaic admixture is only bound to go up.

I will post the abstract, and any further comments specific to the paper itself as an update here.

UPDATE:

From the paper:
Interestingly, the Mbuti represent the only population in our survey that carries the introgressive variant at all three candidate loci, despite the fact that no Mbuti were represented in our initial sequencing survey. Given that the Mbuti population is known to be relatively isolated from other Pygmy and neighboring non-Pygmy populations (26), this suggests that central Africa may have been the homeland of a nowextinct archaic form that hybridized with modern humans.
...
The emerging geographic pattern ofunusual variants discovered here suggests that one such introgressionevent may have taken place in central Africa (wherethere is a very poor fossil record). Interestingly, recent studiesattest to the existence of Late Stone Age human remains witharchaic features in Nigeria (Iwo Eleru) and the DemocraticRepublic of Congo (Ishango) (30–32). The observation that populations from many parts of the world, including Africa, showevidence of introgression of archaic variants (6, 16, 19) suggeststhat genetic exchange between morphologically divergent formsmay be a common feature of human evolution.
The most interesting thing, to me at least, is not that Africans too admixed with archaic humans, but rather the time depth of the separation of the admixing groups: 0.7My, which contrasts with the 270-440ky estimated for the modern-Neandertal split by Green et al. So, it appears that archaic admixture may extend even beyond the H. heidelbergensis clade which is ancestral to modern humans and Neandertals, and may encompass late H. erectus populations.

We are not in classical multiregional territory quite yet (the multiregional model posits a single interbreeding species within the Homo genus throughout its history, consisting of multible regional populations), but Hammer et al. have certainly joined the Neandertal/Denisovan papers into showing that the central prediction of multiregional evolution, namely that there was no reproductive isolation between Homo populations long before the advent of "anatomically modern" humans.

Of course it should be noted that we should not necessarily envision direct intermixture between sapiens and erectus populations. The archaic segments may well have been transmitted via an intermediate erectus-admixed form to the sapiens population.

UPDATE II: A poster at dna-forums makes a rather good point:
So now we are required to believe that homo sapiens interbred first in Eurasia (and the north of Africa) and only afterwards with their closest African neighbors?
It is certainly counterintuitive that admixture with archaics would have happened in Africa after it happened in Eurasia. The latter event was initially theorized to coincide with the first exit of anatomically modern humans to the Levant more than 100,000 years ago, but has been lately redated to 37-86ky BP.

It is difficult to believe that Homo sapiens waited 160ky to mix with his archaic neighbors in Africa (=the period between Omo and 35ky BP) and yet started hooking up right away with Eurasian archaics.

PNAS doi: 10.1073/pnas.1109300108

Genetic evidence for archaic admixture in Africa

Michael F. Hammer et al.

A long-debated question concerns the fate of archaic forms of the genus Homo: did they go extinct without interbreeding with anatomically modern humans, or are their genes present in contemporary populations? This question is typically focused on the genetic contribution of archaic forms outside of Africa. Here we use DNA sequence data gathered from 61 noncoding autosomal regions in a sample of three sub-Saharan African populations (Mandenka, Biaka, and San) to test models of African archaic admixture. We use two complementary approximate-likelihood approaches and a model of human evolution that involves recent population structure, with and without gene flow from an archaic population. Extensive simulation results reject the null model of no admixture and allow us to infer that contemporary African populations contain a small proportion of genetic material (≈2%) that introgressed ≈35 kya from an archaic population that split from the ancestors of anatomically modern humans ≈700 kya. Three candidate regions showing deep haplotype divergence, unusual patterns of linkage disequilibrium, and small basal clade size are identified and the distributions of introgressive haplotypes surveyed in a sample of populations from across sub-Saharan Africa. One candidate locus with an unusual segment of DNA that extends for >31 kb on chromosome 4 seems to have introgressed into modern Africans from a now-extinct taxon that may have lived in central Africa. Taken together our results suggest that polymorphisms present in extant populations introgressed via relatively recent interbreeding with hominin forms that diverged from the ancestors of modern humans in the Lower-Middle Pleistocene.

Link

Related:

14 comments:

aeolius said...

Slightly off topic but foynd through a back search.
In 2009 you quote Herodotus as seeing great similarity between Ethiopians and Indians. If he used Ethiopian as a generic for African then so be it. But if he is using it as Horn of Africans then I do wonder
if there is a comparison with that population and the Onge?
Thanks

Cheryl Cayemberg said...

Just read an article on this today. Very interesting indeed!

Dienekes said...

There is no relationship between Ethiopians and Onge that I can see.

Xaver said...

To your point 7: Indeed, unquestionably modern, but archaic-looking forms persist in the available record down to a few thousand years ago, such as the Fish Hoek and Boskop samples.

According to John Hawks, Boskop never existed as a hominid species, a theory obsolete for 50 years.
Has anything new arisen in recent time?

Dienekes said...

I am not sure what you are arguing for. Boskop is well-known (see "Fossil evidence for the origin of Homo sapiens" in the related links for a description of its non-modern character).

terryt said...

"The observation that populations from many parts of the world, including Africa, showevidence of introgression of archaic variants (6, 16, 19) suggeststhat genetic exchange between morphologically divergent formsmay be a common feature of human evolution".

And I would suggest that it is a common feature in the evolution of virtually all species. It would certainly explain the apparent rapid change in species in the 'punctuated equilibrium' model.

Eze said...

Interesting stuff!

I wish Hammer et al. examined more African groups though. I'll predict that non-Bantu East Africans have the least Archaic admixture in Sub-Saharan Africa.

Onur said...

It is certainly counterintuitive that admixture with archaics would have happened in Africa after it happened in Eurasia. The latter event was initially theorized to coincide with the first exit of anatomically modern humans to the Levant more than 100,000 years ago, but has been lately redated to 37-86ky BP.

It is difficult to believe that Homo sapiens waited 160ky to mix with his archaic neighbors in Africa (=the period between Omo and 35ky BP) and yet started hooking up right away with Eurasian archaics.


I agree. But, unfortunately, due to the effects of recombination and drift on genomes, which you summarized above, we may never find evidence of so ancient archaic admixtures in modern humans without appealing to aDNA analyses on archaic human fossils.

South Central Haplo said...

Lot of groups spread from South Asia. What is the level of admixture in South Asians?.

Onur said...

We must note that detection of archaic admixture is a bit like detective work.

Maybe Michael Hammer is really the Mike Hammer of human genetics after all.

Seinundzeit said...

"We are not in classical multiregional territory quite yet (the multiregional model posits a single interbreeding species within the Homo genus throughout its history, consisting of multiple regional populations), but Hammer et al. have certainly joined the Neandertal/Denisovan papers into showing that the central prediction of multiregional evolution, namely that there was no reproductive isolation between Homo populations long before the advent of anatomically modern humans."

This is a bit muddled. If there was no reproductive isolation, and "We think there were probably thousands of interbreeding events...It happened relatively extensively and regularly." (Hammer et al.), then how did these populations not belong to the same species? If this is not classic "Multi-Regionalism", than what is?

terryt said...

"Maybe Michael Hammer is really the Mike Hammer of human genetics after all".

He has certainly done some great work. Even as far back as 1995 he published a paper on the Japanese:

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1801189/

Still a useful paper.

"If this is not classic 'Multi-Regionalism', than what is?"

I'm very much inclined to agree.

Onur said...

We need a fresh new and Darwinian approach to the classification of organisms. We should abandon the concept of taxonomic ranks, which is pre-Darwinian and unscientific, and should define organisms completely genealogically (clade-based) from bottom to top without artificially lumping them into a hierarchy of fixed ranks. PhyloCode is a good initiative in this regard:

http://en.wikipedia.org/wiki/PhyloCode

Anon9GAva4Ez said...

This is a bit muddled. If there was no reproductive isolation, and "We think there were probably thousands of interbreeding events...It happened relatively extensively and regularly." (Hammer et al.), then how did these populations not belong to the same species? If this is not classic "Multi-Regionalism", than what is?

Thousands of interbreeding events are nothing compared to the bazillions of endospecific breeding events. That's why "some admixture" differs from regional continuity. I find weird how these new findings on recent years have been making many people assume some sort of nearly strawman-esque ultra-hard version of out-of-Africa and a "one drop rule" version of multiregional evolution at the same time. 2-4% of archaic admixture implies 98-96% of out-of-africa.

If we were really reading news in favor of multiregional evolution, they would have to be in the lines of "it turns out that 98-96% (hell, I give 50-30% for a decent but yet far range that really deserves the multiregional label) of the genetic pools of people from the different continents is actually a continuous heritage from erectine populations from each continent, and only 2-4% came from more recent migrations out of Africa -- and we had it totally wrong all along".



All the multiregionalist hype over these interbreeding events remind me of when someone is saying something about how there is much horizontal gene transfer between the branches on the tree of life "and therefore Darwin got it all wrong", or something about epigenetics and "Lamarck got it right after all", and things like that.

I find much more weird and "hype-worthy" things like the fact that it seems that multirregional continuity does not seem to hold not even for later times. How archaic sapiens -- not erectus -- from each place look much more like each other than to their present-day counterparts, and likewise, with present-day populations as well.

http://dienekes.blogspot.com/2011/03/out-of-south-africa-out-of-anywhere.html

http://dienekes.blogspot.com/2007/01/new-kostenki-finds-in-science-magazine.html

It's like how it's an interesting question why there are no more enantiornithes today, but all the time is wasted dealing with some bullcrap about some stupid "evidence against" dinosaurs being the ancestors of birds, trying to jump to the conclusion that they're basal archosaurs/thera


PS.: I think it's a bit hurried to say that "there was no reproductive isolation". These numbers, couldn't they be the same even if, hypothetically, the robertsonian translocation occurred only just about the rise of modern Homo sapiens? I think that something like would make everything much less odd than it is. I'd normally expect something much more on the direction of those weaker multiregional evolution models, not something so close to a ultra-strong out of Africa and the unstoppable march of the pure real humans, as the things are. A biological barrier such as the robertsonian translocation perhaps gives explains "everything", how modern humans are the "odd man out" anatomically, and how we're so relatively unmixed, at the same time it's a somewhat leaky isolation that allows for some interbreeding.