May 15, 2010

Neanderthals had ancestral Microcephalin gene

Evans et al. had previously proposed that the derived MCPH1 allele had entered the modern human gene pool by Neandertals. This paper tests this by looking at a 50,000 year BP Neandertal, finding them to have the ancestral allele, thus weakening the case for introgression.


The authors write:
Evans and colleagues proposed that haplogroup D originated from a lineage separated from modern humans for 1.1 million years and introgressed into the human gene pool by 37,000 years ago, probably from a Neanderthal stock [4].

However, simulation approaches have shown that two key hypotheses of this model, namely positive selection and admixture, can be relaxed as long as Eurasia was settled from an African population that was both subdivided and under expansion [5]. Interestingly, variation in neurocranial geometry have recently suggested significant levels of geographic structure among early modern humans from Africa [6]. In addition, no direct empirical evidence supports a third key component of the model, ie. that Neanderthals carried alleles of the D haplogroup.
This is a great paper in view of the recent controversy about modern human-Neandertal admixture. I have been on the skeptic camp, and this certainly strengthens my African structure thesis.

Here is how the authors try to explain the deep divergence between the ancestral and derived haplotypes:
There is a broad agreement that the contribution of archaic Homo populations to the modern gene pool, if any, must have been very limited [33], [34]. Different lines of evidence concur to suggest that the dispersal of anatomically modern humans from Africa was accompanied by repeated founder effects [35]–[38]. If these founder effects were drastic, most or all gene genealogies should actually be shallow, and hence the occurrence of ancient splits would imply some degree of introgression from archaic human forms. However, different consequences would be expected if only mild founder effects occurred when anatomically modern humans moved out of Africa. Under these conditions, gene trees would have a strong random component, and a certain fraction thereof, even in the absence of selection, would show two highly divergent major lineages [39]. The likelihood of finding gene genealogies with a very old common ancestor and very differentiated lineages would be even higher if the source African population was subdivided and structured genetically before dispersal, which is what most studies clearly suggest [40]–[43]. These theoretical considerations are actually matched by consistent results in simulation studies [5], [34], [44] and by variation in neurocranial geometry, suggesting significant levels of geographic structure among early modern humans from Africa [6].
One more piece of evidence in favor of the idea that seemingly archaic DNA in modern Eurasians is not due to the more fashionable "Neandertal introgression" but to ancient African population structure.

PLoS ONE 10.1371/journal.pone.0010648

The Microcephalin Ancestral Allele in a Neanderthal Individual

Martina Lari et al.

Background
The high frequency (around 0.70 worlwide) and the relatively young age (between 14,000 and 62,000 years) of a derived group of haplotypes, haplogroup D, at the microcephalin (MCPH1) locus led to the proposal that haplogroup D originated in a human lineage that separated from modern humans >1 million years ago, evolved under strong positive selection, and passed into the human gene pool by an episode of admixture circa 37,000 years ago. The geographic distribution of haplogroup D, with marked differences between Africa and Eurasia, suggested that the archaic human form admixing with anatomically modern humans might have been Neanderthal.

Methodology/Principal Findings
Here we report the first PCR amplification and high- throughput sequencing of nuclear DNA at the microcephalin (MCPH1) locus from Neanderthal individual from Mezzena Rockshelter (Monti Lessini, Italy). We show that a well-preserved Neanderthal fossil dated at approximately 50,000 years B.P., was homozygous for the ancestral, non-D, allele. The high yield of Neanderthal mtDNA sequences of the studied specimen, the pattern of nucleotide misincorporation among sequences consistent with post-mortem DNA damage and an accurate control of the MCPH1 alleles in all personnel that manipulated the sample, make it extremely unlikely that this result might reflect modern DNA contamination.

Conclusions/Significance
The MCPH1 genotype of the Monti Lessini (MLS) Neanderthal does not prove that there was no interbreeding between anatomically archaic and modern humans in Europe, but certainly shows that speculations on a possible Neanderthal origin of what is now the most common MCPH1 haplogroup are not supported by empirical evidence from ancient DNA.

Link

30 comments:

Maria Lluïsa said...

shared by neanderthals and non-africans are apparently neutral or have and unknown function, but clearly, microcephalin is not among them.
This study proves that neanderthals are not the source of microcephalin. This is not so surprising, because we never had evidence of that, only speculations. But by no means proves that these neutral regions shared with non africans are the result of an structure population within Africa, which is actually quite unlikely. Moreover, one of the scientits that investigated the MC1R version in neanderthals, Carles Lalueza, was against the idea of interbreeding until now, meaning that if they claim there was interbreeding, it's because they have find strong evidence for it.

Average Joe said...

Does this mean that ancient humans were racially divided?

belenos said...

I'm not a geneticist, but is a dna test on one individual not very inconclusive evidence?

If we found it in one Neanderthal individual it would be very strong (near conclusove) evidence of interbreeding, but its absence proves nothing.

Dienekes said...

shared by neanderthals and non-africans are apparently neutral or have and unknown function, but clearly, microcephalin is not among them.

No selective advantage for the derived MCPH1 allele has been discovered so far, and it's not due to lack of trying to discover it. So, I'm not counting on the selection hypothesis at this point.

But by no means proves that these neutral regions shared with non africans are the result of an structure population within Africa, which is actually quite unlikely.

Why do you think it's unlikely?

If we found it in one Neanderthal individual it would be very strong (near conclusove) evidence of interbreeding, but its absence proves nothing.

You are correct that it "proves" nothing, but it changes the probabilities.

Brent said...

Abit OT, but Dienekes, what did you mean by the following on Frost's recent blog post?

"There were actually other archaic humans in Africa itself, and even subdivision with anatomically modern humans."

What other archaic humans are you reffering to, and what exactly is that paper arguing for?

Maria Lluïsa said...

"No selective advantage for the derived MCPH1 allele has been discovered so far, and it's not due to lack of trying to discover it. So, I'm not counting on the selection hypothesis at this point."
I have read that it has a role in brain development, and due to its high levels in eurasians it has been suggested that it might have been positive selected. Anyway, it was only an speculation trying to justify its huge levels outside Africa.
"Why do you think it's unlikely?"
Beucause no africans have been discovered with these 1-4% neandertal, only non-africans; and also because if this ancient structure within Africa is so likely, the authors wouldn't have claimed for the interbreeding. Keep in mind that if this study (neanderthal genomics) is invalidated, then the authors may be ridiculized, because the press is full of comments like "they're not extinct, they're us".
What I found very surprising is that most of these anthropologists/genetics were against the idea of interbreeding before publishing the study. Therefore, they might have find some important evidence to believe in it.
"You are correct that it "proves" nothing, but it changes the probabilities."
Well, that neanderthals aren't the source of microcephalin is something we know since 2006. The Max Planck genetists didn't have find the derived allele in none of the 3-6 neanderthal genomes.
There are other candidate alleles, for example H2 tau haplotype. Its age was calculated at more than 2.000.000, and the fact that it appears to be european specific makes it to be a perfect candidate to prove its neandertal origins. But a recent study changed these dates to less than 50.000 years. The autors concluded that it is quite unlikely to be originated in neanderthals.
I don't know about microcephalin, but I don't see any relation between it and neanderthal interbreeding. It appears to be an exclusive modern allele that appeared about 37.000 years ago, when neanderthals were nearly extinct.

catperson said...

I'm confused by your African structure hypothesis. Are you arguing that neanderthals entered Africa and breeded with a subset of Africans, and that subset went on to evolve into Eurasians, thus genetic link between Eurasians and Africans. If so, I don't see the significance, since you agree that neanderthals mixed with some humans, you just disagree about where & when it took place.

Or are you saying (as I since have been arguing) that neanderthal genes mutated into the ancestral hominid line before humans and neanderthals diverged and they were only selected for in neanderthals and human Eurasians because they were advantageous outside Africa.

eurologist said...

The recent science paper By Green at al. actually found the same thing: none of their Neanderthal samples had the derived (Haplogroup D) microcephalin variant.

I have mentioned before several times that if you find some ancient-looking stretches in Eurasian DNA, chances are, if you look hard enough, you will also find them in Africa. However, the Green at al. findings are very, very different from that, in that the stretches are derives and shared by Neanderthals. I really don't see how the microcephalin genes can be used as an argument against gene flow from Neanderthals to modrrn humans, when (i) the Neanderthal samples used are relatively old and when (ii)they did not find gene flow from AMHs to Neanderthals, anyway.

Dienekes said...

What other archaic humans are you reffering to, and what exactly is that paper arguing for?

Homo originated in Africa and spread into Eurasia where it developed into other forms (such as Neandertals and "hobbits"). Later, Homo sapiens replaced all pre-existing Homo types in Eurasia. I don't see any a priori reason why it did not do the same thing in Africa, i.e., it replaced non-sapiens Homo in Africa itself. I simply don't see how the entirety of Homo would undergo a bottleneck in Africa that would leave only Homo sapiens as the single occupant of the continent. There are a couple reasons for this: first, Africa is the cradle of Homo, and I don't really see any reason for a serious pan-African bottleneck there. Second, Africa is environmentally varied because of its latitudinal orientation, so any climatic deterioration there would still not leave a serious bottleneck. Third, we have good evidence of structure in Africa today, despite much greater mobility that has blurred lines, so I don't see why such structure could not have existed in the past.

In short, I think that Homo sapiens emerged in East Africa or thereabouts, that it was reproductively (semi-)isolated from other African Homo populations and that it spread both Out of Africa and Deeper Into Africa, absorbing pre-existing Homo populations there. Even at the onset of modernity there is evidence for variation in Africa, and e.g., Omo is more "modern" than Homo sapiens idaltu depsite the fact that the latter is a later specimen, and Kabwe is more archaic than both. So, there is good evidence, I think for substantial variation of African homo, and no real reason to think that Homo sapiens spread into empty territory when it populated the continent.

Beucause no africans have been discovered with these 1-4% neandertal, only non-africans

Well, the authors didn't really look at native East Africans, so, until they do, they have no real evidence that what they found is limited to non-Africans.

Are you arguing that neanderthals entered Africa and breeded with a subset of Africans, and that subset went on to evolve into Eurasians, thus genetic link between Eurasians and Africans.

No, I am arguing that Homo sapiens admixed with non-sapiens Homo in Africa itself, and also that Homo sapiens -> Neandertal admixture rather than the reverse can explain the authors' findings, because the authors make an implicit assumption about the relationship of all ancient Homo sapiens that may have contributed genes to Neandertals, namely that they have a similar relative distance to Yoruba and San.

However, the Green at al. findings are very, very different from that, in that the stretches are derives and shared by Neanderthals. I really don't see how the microcephalin genes can be used as an argument against gene flow from Neanderthals to modrrn humans, when (i) the Neanderthal samples used are relatively old and when (ii)they did not find gene flow from AMHs to Neanderthals, anyway.

That is not a problem if the derived stretches are descended from the common ancestor of Homo sapiens and Neandertals or they entered the Neandertal gene pool by admixture with modern humans, which the authors can't really reject without extra assumptions. Your contention (ii) is wrong, because they did not study gene flow from AMHs to Neandertals, and can only exclude modern Eurasian-like admixture in Neandertals: but there is no reason to think that e.g., the anatomically modern humans that were in the Levant 90ky ago were like modern Eurasians.

Maria Lluïsa said...

"Well, the authors didn't really look at native East Africans, so, until they do, they have no real evidence that what they found is limited to non-Africans"

At first I was surprised to find that they only have analyzed the genome of 5 modern humans, but apparently more africans and non-africans have been analyzed. I could be wrong, but this page (http://genome.ucsc.edu/cgi-bin/hgTrackUi?db=hg18&c=chr10&position=chr10:69713986-69714099&g=ntOoaHaplo) says:

"Green et al. used 1,263,750 Perlegen Class A SNPs, identified in 71 individuals of diverse ancestry (see Hinds et al.), to identify 13 candidate gene flow regions (Supplemental Online Materials Text 17). 24 individuals of European ancestry and 24 individuals of Han Chinese ancestry were used to represent the non-African population, and the remaining 23 individuals, of African American ancestry, were used to represent the African population."
If among these african americans there are individuals with east-african ancestry (which is quite likely) the authors might have been observed somewhat different results.
The results can also be explained if you assume that all modern humans originated from a pre-neandertal eurasian population and that a group migrated to Africa 80.000 years ago. This would explain why non-africans are a bit more closely related to neandertals, but I still find it to be quite unlikely since the oldest mtDNA/ Y chromosome clades are found within Africa.

catperson said...

No, I am arguing that Homo sapiens admixed with non-sapiens Homo in Africa itself, and also that Homo sapiens -> Neandertal admixture rather than the reverse can explain the authors' findings, because the authors make an implicit assumption about the relationship of all ancient Homo sapiens that may have contributed genes to Neandertals, namely that they have a similar relative distance to Yoruba and San.

So if homo sapiens mixed with a non-sapien form of homo (any one in particular?) in Africa, how does that explain why some modern homosapiens are more genetically similar to neanderthals? Did only the African homo sapiens who would later leave Africa mix with the non-sapien homo, and did the hybrids of the sapien & non-sapien homo leave Africa too (admixing with neanderthals) giving neanderthals the genes of the subset of sapiens who would become Eurasians? Doesn't adding a third group to the equation (non-sapien homo) violate Occam's Razor?

Bottom line: Do you believe the strong form of the Out of Africa (total replacement) hypothesis still stands?

Dienekes said...

how does that explain why some modern homosapiens are more genetically similar to neanderthals?

Because Neandertals and Homo sapiens share a common ancestor within the Homo lineage.

Alternatively, because Neandertals have Homo sapiens admixture.

Certainly I can't dismiss the possibility of Neandertal->sapiens admixture, but I don't see any reason to think this is a better explanation for the observed data.

Now, if I saw this "Neandertal ancestry" occurred primarily in Caucasoids, i.e., people from West Asia and Europe where any hypothetical admixture took place, Neandertals would be the prime candidate for this admixture. Since I don't see that, I see no reason to dismiss the other explanations.

Doesn't adding a third group to the equation (non-sapien homo) violate Occam's Razor?

I am not adding a third group to the equation. There is evidence for non-sapiens Homo in Africa, and even if there wasn't, there is no reason to think that archaic Homo was wiped out in Africa and his descendants survived in Eurasia. So, what I am saying is that Homo sapiens-archaic admixture in Africa is every bit as good an explanation as Homo sapiens-Neandertal admixture in Eurasia, and IMO has the added bonus that it has no difficulty in explaining the uniformity of "Neandertal" admixture across Eurasia.

Dienekes said...

If among these african americans there are individuals with east-african ancestry (which is quite likely)

I see no evidence that it is "quite likely" as African Americans in published studies that compared them to African populations are overwhelmingly related to West African groups.

Brent said...

Actually, another question- what other homo species are you saying lived in africa back then?

catperson said...

Dienekes, let me try to summarize your argument to see if I comprehend: Both anatomically modern humans (AMH) and Neanderthals evolved from an archaic form of homo that lived in Africa. The subset of anatomically modern humans who would later leave Africa (proto-non-Africans) mated with the archaic homo and this caused them to become more genetically similar to Neanderthals because either (a) they absorbed archaic homo genes which are related to Neanderthal genes since archaics are the last common ancestor of AMH and Neanderthals, and/or, (b) archaics absorbed genes from proto-non-Africans which Neanderthals then absorbed from mating with the hybridized archaics; thus causing a genetic affinity between Neanderthals and non-African AMH.

Dienekes said...

Dienekes, let me try to summarize your argument to see if I comprehend: Both anatomically modern humans (AMH) and Neanderthals evolved from an archaic form of homo that lived in Africa.


Correct. This archaic form of Homo was, however, part of a structured African population, i.e., there were other types of people in Africa itself.

The subset of anatomically modern humans who would later leave Africa (proto-non-Africans) mated with the archaic homo and this caused them to become more genetically similar to Neanderthals


No, the people who left Africa did not mate with any archaic Homo, as they passed from east Africa into West Asia. It is the people who went deeper INTO Africa that eventually encountered the archaic populations of Homo that made them more DISSIMILAR to Neandertals.

An alternative explanation (which does not require African structure) is if modern humans who left Africa early contributed genes TO Neandertals during their co-existence in the Near East. Actually I find this scenario to be likely as well, as we have clear evidence of AMHs in the Near East ~100ky before the main migration wave ~60-40ky.

catperson said...

No, the people who left Africa did not mate with any archaic Homo, as they passed from east Africa into West Asia. It is the people who went deeper INTO Africa that eventually encountered the archaic populations of Homo that made them more DISSIMILAR to Neandertals.

Why would mating with archiac Homo make modern humans more dissimilar to neanderthals? Did not both neanderthals and modern humans evolved from this archaic homo? Thus, by mating with the archaic homo, one is mating with the common ancestor one shares with neanderthals and thus increasing (or at least maintaining) the genetic link with neanderthals. Or are you saying they mated with some archaic homo that was not ancestral to neanderthals?

An alternative explanation (which does not require African structure) is if modern humans who left Africa early contributed genes TO Neandertals during their co-existence in the Near East. Actually I find this scenario to be likely as well, as we have clear evidence of AMHs in the Near East ~100ky before the main migration wave ~60-40ky.

And these early AMH in the Near East died out I believe. I think an earlier mating is more plausible, because 100 K ago, humans and neanderthals may still have been sub-species within the same species, but by 60 K ago, the two groups may have been isolated long enough that they could no longer breed.

But if humans and neanderthals did breed 100 K ago, would not neanderthals have left genes in humans too. But then it wouldn't matter since that early group died out & with them died the neanderthal genes.

Andrew Oh-Willeke said...

Dienekes:

Your argument is making a lot more sense when spelled out in the comments.

Introgression to Neanderthal from an abortive wave of modern humans to the Near East before the main migration wave is plausible (particularly for a Neaderthal population reasonably close to the Near East about 35,000 years after wave -1 ended). And, the Southern Route of migration to Asia doesn't provide a lot of time or space for Neaderthal and early modern human admixture.

As I understand this theory, the assumption would be that there would be little Neaderthal to modern human introgression in either wave (hence little need to show why one period was different from the other).

A big plus of this approach is that is reconciles the lack of any sign of any Neanderthal Y-DNA or mtDNA lineages in Out of Africa modern humans (as shown by relatively recent divergence dates of all such lineages from their African sources), despite an apparent 1%-4% aDNA admixture.

If the event that Neanderthal effective population sizes were very small and stable or declining, and that this population had at least three regional subpopulations is right, and the one the samples come from would be closest to the modern human population is right, it wouldn't take many cases of admixture to get a 4% modern human component into the regional Neanderthal mix, particularly if the modern human component provided selective advantages. You would need something on the order of 30-60 hybrid child events over 1,000 years plus of cohabitation, which is on the order of one per generation, and less with selective advantages.

You would think that any admixture big enough to leave a 4% aDNA admixture would leave at least a handful of relict Y-DNA and mtDNA lineages in European or Near Eastern populations, perhaps not 4%, but surely 0.04% or more, as even selectively disadvantageous mtDNA or Y-DNA probably wouldn't be selectively disadvantageous in all circumstances.

Yet, the Khoisan/Pygmy to other African break is to only about 70,000 years ago, and the African to non-African breaks are more recent, and those division dates can be off by a very great deal and still not make a case for being Neanderthal lineages, and there are parallel Y-DNA dates. Yet, not a single mtDNA lineage pre-L, or a single Y-DNA lineage of comparable phylogenetic distance has emerged out of thousands of very diverse samples.

The notion that Africa had pre-modern humans whom modern humans replaced and/or partially absorbed via admixture in a "Deeper into Africa" expansion also has merit.

Of course, largely one way gene flow into Neaderthal populations makes African structure (where presumably gene flow would also be largely one way) irrelevant.

Nate said...

How do know that H. erectus didn't expand into Europe some time, perhaps 1 million years ago and from there evolve into H. heidelbergensis or the like and THEN expand into Africa some 500,000 KYA to eventually become H. sapiens. In this scenario, H. neanderthalensis envolved in situ in Europe.

Based on the fossil evidence, I'm not sure there's necessarily a reason to assume an all-African lineage to modern humanity.

Brent said...

But Dienekes, just what other archaic homo populations existed in africa at this time?

Gioiello said...

ITALIAN GENE STUDY WEAKENS NEANDERTHAL-MODERN EUROPEAN LINK.412 words17 May 201020:08ANSA - English Media ServiceANSAENEnglish
(ANSA) - Florence, May 17 - An Italian study out this week lends fresh weight to the theory that today's Europeans are not descended from Neanderthals, despite a recent study indicating Neanderthal DNA is common in many modern humans.The research, which appears in the international Public Library of Science journal, compared modern human DNA with that of Neanderthal man, focusing on a gene responsible for a condition known as microcephaly, microcephelin.A variant of the gene appears in a specific genetic grouping known as Haplogroup D, which is one of the categorizations used by scientists to map the early migrations of population groups.Haplogroup D appeared around 37,000 years ago and is now common throughout the world but is extremely rare in Sub-Saharan Africa."Until now, this particular distribution had been interpreted as evidence that Haplogroup D had originated with Neanderthals," explained a statement by Florence University, which produced
the study in collaboration with the universities of Siena and Ferrara, as well as the Milan Institute of Biomedical Technologies and the Verona Natural History Museum."It was hypothesized that this variant had then been incorporated into the genome of modern humans by admixing between Neanderthals and the ancestors of modern Europeans and Asians, after they had already left Africa".However, the Italian experts, who were joined by scientists from Marseilles and Lyons universities for the research, have now proved this was not the case.The sequencing of the Neanderthal genome indicates Neanderthals did not have the Haplogroup D gene and are no genetically closer to Europeans than modern Africans."The study does not prove there was no mixing between different human forms in Europe," the study's coordinator David Caramelli, acknowledged."However, it certainly shows that speculation over a possible Neanderthal origin for the most widespread variant of the
microcephalin gene in European populations is not supported by evidence obtained from ancient DNA."In other words, we cannot exclude that a tiny fraction of our DNA is of Neanderthal origin but we can say for certain that this was not the case with microcephalin".The study appears just days after research published in Science grabbed worldwide headlines after concluding that many modern humans possess some Neanderthal ancestry.Overturning previously held beliefs that Neanderthals had made little or no contribution to our inheritance, the study suggested that between 1% and 4% of modern European and Asian genomes appeared to have come from Neanderthals.

terryt said...

"How do know that H. erectus didn't expand into Europe some time, perhaps 1 million years ago and from there evolve into H. heidelbergensis or the like and THEN expand into Africa some 500,000 KYA to eventually become H. sapiens. In this scenario, H. neanderthalensis envolved in situ in Europe".

I strongly suspect that's exactly what happened.

eurologist said...

How can this study put the Greene et al. paper in a different light or "weaken the Neanderthal link" when the Greene paper actually found the same result concerning the microcephalin gene?

Dienekes said...

eurologist, I am quoting Evans et al:

"Under this scenario, the two populations are reproductively isolated for a prolonged period such that one population was fixed for the D allele, whereas the other population was fixed for the non-D allele. The two populations were then thoroughly admixed ≈37,000 years ago, and the high frequency of the D allele in the final admixed population is not the result of selection but rather because the D-bearing population contributed a significant faction to the admixture."

What the Neandertal studies show is that Neandertals were clearly not fixed for the derived allele, and hence the Evans et al. model does not match reality.

Note that Evans et al. reject the scenario that the D allele persisted for a long time as a balanced polymorphism. So, if the D allele arose on the Neandertal line it would have been fixed (with very high probability) in a million years, because the non-D allele could not have been maintained by balancing selection.

Thus, we have a model in which the D allele was fixed in Neandertals and entered the modern gene pool by admixture. The model is falsified (or at least much much less probable) by the ancient DNA evidence.

So we didn't get the D allele from Neandertals.

eurologist said...

So we didn't get the D allele from Neandertals

Of course not. All I am saying is that Green et al. - who are proposing 1% to 4% Neanderthal admixture into Eurasians - got the same result - and it didn't bother them with respect to their main conclusion:

"This analysis shows that some old haplotypes most likely owe their presence in present-day non-Africans to gene flow from Neandertals. However, not all old haplotypes in non-Africans may have such an origin. For example, it has been suggested that the H2 haplotype on chromosome 17 and the D haplotype of the microcephalin gene were contributed by Neandertals to present-day non-Africans (12, 79, 80). This is not supported by the current data because the Neandertals analyzed do not carry these haplotypes."

So they recognize the fact that there is other structure, and I think most people agree that both the archaeological and genetic evidence shows that there was (and still is) significant substructure in Africa. I mean, clearly the San show that populations in Africa can remain largely isolated for hundreds of thousands of years.

How do know that H. erectus didn't expand into Europe some time, perhaps 1 million years ago and from there evolve into H. heidelbergensis or the like and THEN expand into Africa some 500,000 KYA to eventually become H. sapiens. In this scenario, H. neanderthalensis envolved in situ in Europe.

Based on the fossil evidence, I'm not sure there's necessarily a reason to assume an all-African lineage to modern humanity.


I wouldn't take it that far - I think the fossil evidence points to that (i) African heidelbergensis (rhodesiensis) and European heidelbergensis were anatomically very similar, (ii) they likely derived from African H. ergaster (and not Asian erectus) over a million years ago, and (iii) there was a wide range of homo approaching AMHs in Africa, while all we know from Europe is that around 300,000 to 400,000 years ago, Heidelbergensis transformed into a single population - Neanderthals. And everything that makes them unique, anatomically, is not shared on the African continent, and is largely not shared by modern humans.

So, as I have mentioned before, I find it possible and even quite likely that over a ~500,000 year period there was some gene flow between heidelbergensis (until perhaps very, very early Neanderthal) and North Africans. But everything points to that this must have largely ceased ~400,000 years ago (with the exception of the Near East event ~100,000 years ago). In fact, it is quite likely that such climate windows only opened every 50,000 to 100,000 years, for a few thousand years, each.

At any rate, the total population Africa could support was always more than an order of magnitude larger than Europe's, and probably 2 or 3 orders of magnitude larger than any groups migrating south. So, it would seem that any such genetic contribution from the North would have been rather limited (while an inverse flow from northward should have persisted, whenever possible). It may still have been important for the evolution of AMHs, though - especially between 800,000 and 500,000 years ago, when Heidelbergensis acquired larger brains and the African tool set became more elaborate.

I think if we accept the notion that Neanderthals derive from heidelbergensis (extremely likely), the fact that Neanderthals are so similar to us genetically supports the idea that there must have been continued and substantial gene flow with Africa some time not too far before their divergence from heidelbergensis.

mathilda said...

I can't see how this study does anything other than show we didn't pick up this mutation from one particular pre AMH colonisation date Neanderthal, Dienekes. It doesn't strenghten the structure in Africa interpretation at all.

I'm still supporting the Neanderthal ancestry, mainly on the grounds that they seem to have been sufficiently close enough for Paabo to have said that they come into the range of modern humans; no major differences means no barrier to hybridisation other than cultural.

"Well, the authors didn't really look at native East Africans, so, until they do, they have no real evidence that what they found is limited to non-Africans."

I assume they didn't use East Africa because of the well known Eurasian contribution to their ancestry, so looking there for shared traits wouldn't really prove anything. Maybe if they did and found it at a lower level than in non-Africans it might support it only being there as baggage from the back migrations .

Dienekes said...

I can't see how this study does anything other than show we didn't pick up this mutation from one particular pre AMH colonisation date Neanderthal, Dienekes. It doesn't strenghten the structure in Africa interpretation at all.

You are not including in your thinking the fact that D diverged from A about a million years ago, and would have been fixed in Neandertals over 40k generations or so. So, the fact that the Neandertals tested so far were A suggests strongly that D did not arise in the Neandertal line.

J. Lyon Layden said...

"There is a broad agreement that the contribution of archaic Homo populations to the modern gene pool, if any, must have been very limited [33], [34]. Different lines of evidence concur to suggest that the dispersal of anatomically modern humans from Africa was accompanied by repeated founder effects [35]–[38]. If these founder effects were drastic, most or all gene genealogies should actually be shallow, and hence the occurrence of ancient splits would imply some degree of introgression from archaic human forms. However, different consequences would be expected if only mild founder effects occurred when anatomically modern humans moved out of Africa. Under these conditions, gene trees would have a strong random component, and a certain fraction thereof, even in the absence of selection, would show two highly divergent major lineages [39]. The likelihood of finding gene genealogies with a very old common ancestor and very differentiated lineages would be even higher if the source African population was subdivided and structured genetically before dispersal, which is what most studies clearly suggest [40]–[43]. These theoretical considerations are actually matched by consistent results in simulation studies [5], [34], [44] and by variation in neurocranial geometry, suggesting significant levels of geographic structure among early modern humans from Africa [6]."

J. Lyon Layden said...

So what? It could have evolved in The Hobbit, Soloensis, Denisovans, Red Deer Cave people. Maybe it was even passed to Balangoda man before it passed to us. No space for my full comment here, so I'm posting on my blog, with a link to yours: http://prehistoricfantasy.blogspot.com/

J. Lyon Layden said...

Mary Luis- Neanderthals were far from extinct 37,000 years ago. We were just barely getting into Europe at that point. The last Neanderthal died in southern Iberia only 21 to 24 thousand years ago- get with the times!
Also, 2 million year old gene that didn't get into us until 50 thousand years ago cannot have come from neanderthals- highest estimates for the Neanderthal-Human split are around 800,000 years....and a 350000 to 450000 year old split is a lot more probable. That's why I never understood why they were looking for the 990,000 year old gene in Neanderthals in the first place. We always new that soloensis was around at that time, even before the discoveries of red deer and The Hobbit...unless of course you were one of those people who likes denial more than obvious.