May 07, 2010

Tales of Neanderthal admixture in modern Eurasians

I was wary of this paper's conclusion as soon as I realized that the authors contended that Europeans and East Asians did not differ significantly in their levels of Neanderthal admixture. You see, Neandertals were absent from East Asia, so there is no reason for East Asians to have such admixture at all, or to have much less of it than West Eurasians do.

Even if we assumed that undifferentiated Eurasians picked up Neandertal admixture in West Asia, the fact would remain that Europeans co-existed with Neandertals after the ancestors of East Asians had moved on towards the Pacific. So, they ought to have more Neandertal admixture -if such admixture ever took place. Indeed, one of the main arguments for introgression of Neandertal genes in modern humans is the long co-existence of modern humans with Neandertals in Europe.

It could be argued that Neandertal admixture was indeed higher in West Eurasians initially, but the difference was evened out by gene flow across Eurasia. This, however, makes no sense, as the history of Eurasians post-Out of Africa was one of genetic differentiation, which suggests barriers to gene flow, and it would be difficult to imagine a scenario in which the Neandertal component would even out in Eurasia across populations from the Atlantic to the Pacific.

We should also note that the present paper's line of genetic evidence is not really complementary to the palaeoanthropological argument for admixture, as that argument is based on observing phenotypic continuities between Neandertals and modern Caucasoids that could not (according to the argument's proponents) be explained by the simple Out of Africa model and/or observing phenotypic traits of Upper Paleolithic Europeans that seem to be shared with Neandertals but not with UP sapiens outside Europe.

In short: those who use skulls to argue Neandertal introgression predict that Europeans should be somehow closer to Neandertals than East Asians are, but this paper fails to find evidence of this.

So what really happened: an alternative hypothesis

The authors do make an important observation: Neandertal genomes were closer to those of modern Eurasians than to modern Africans. This is an important finding that is incompatible with pure Out-of-Africa. But, Neandertal admixture is not the only way to explain the data.

There is an alternative explanation. It involves the emergence of Homo sapiens and Homo neanderthalensis from a common ancestor and the subsequent admixture of Homo sapiens with populations that have branched out before this divergence. This would account for increased similarity between Eurasians and Neandertals, but without the problem of explaining how "Neandertal" ancestry is so similar in Europeans and East Asians.

What about Africans? Why do they stand further away from Neandertals? The answer is simple: low-level of admixture with archaic humans in Africa itself. It is fairly clear to me that the sapiens line whose earliest examples are in East/South Africa must have been an offshoot of an older African set of populations. We are lucky that Neandertals lived in a climate conducive to bone (or even DNA) preservation, while the African populations inhabiting the tropics left no traces of their existence.

In conclusion: I am not at all convinced that the authors have uncovered evidence of Neanderthal admixture in Eurasians; the alternative explanation is that modern humans and Neandertals were related, modern humans spread from East Africa/West Asia and as they entered deeper into Africa, they interacted with archaic human populations there.

PS: The only potential argument in favor of the authors' hypothesis is the following:
Non-Africans haplotypes match Neandertals unexpectedly often. An alternative approach to detect gene flow from Neandertals into modern humans is to focus on patterns of variation in present-day humans—blinded to information from the Neandertal genome—in order to identify regions that are the strongest candidates for being derived from Neandertals. If these candidate regions match the Neandertals at a higher rate than is expected by chance, this provides additional evidence for gene flow from Neandertals into modern humans.
We thus identified regions in which there is considerably more diversity outside Africa than inside Africa, as might be expected in regions that have experienced gene flow from Neandertals to non-Africans. We used 1,263,750 Perlegen Class A SNPs, identified in individuals of diverse ancestry (78), and found 13 candidate regions of Neandertal ancestry (SOM Text 17). A prediction of Neandertal-to-modern human gene flow is that DNA sequences that entered the human gene pool from Neandertals will tend to match Neandertal more often than their frequency in the present-day human population. To test this prediction, we identified 166 "tag SNPs" that separate 12 of the haplotype clades in non-Africans (OOA) from the cosmopolitan haplotype clades shared between Africans and non-Africans (COS) and for which we had data from the Neandertals. Overall, the Neandertals match the deep clade unique to non- Africans at 133 of the 166 tag SNPs, and 10 of the 12 regions where tag SNPs occur show an excess of OOA over COS sites. Given that the OOA alleles occur at a frequency of much less than 50% in non-Africans (average of 13%, and all less than 30%) (Table 5), the fact that the candidate regions match the Neandertals in 10 of 12 cases (P = 0.019) suggests that they largely derive from Neandertals. The proportion of matches is also larger than can be explained by contamination, even if all Neandertal data were composed of present-day non-African DNA (P = 0.0025) (SOM Text 17).
The problem with the above analysis is in the underlined portion. The deep clade is rather unique to non-Americans of African ancestry, as per reference (78), i.e., a very limited sample of Africans. In short, there is no reason to believe that the identified deep clades matching the Neandertals are really unique to non-Africans, and the pattern can be easily explained by geographical structure within Africa itself.

Related: for a contrary view see John Hawks.

UPDATE: To their credit, the authors consider the scenario I am advocating it here, labeling it Scenario 4. They write:
Although gene flow from Neandertals into modern humans when they first left sub-Saharan Africa seems to be the most parsimonious model compatible with the current data, other scenarios are also possible. For example, we cannot currently rule out a scenario in which the ancestral population of present-day non-Africans was more closely related to Neandertals than the ancestral population of present-day Africans due to ancient substructure within Africa (Fig. 6). If after the divergence of Neandertals there was incomplete genetic homogenization between what were to become the ancestors of non-Africans and Africans, present-day non-Africans would be more closely related to Neandertals than are Africans. In fact, old population substructure in Africa has been suggested based on genetic (81) as well as paleontological data (86).

Whether their model is more parsimonious than Scenario 4 is up to the reader. We know that there is population structure in Africa today, and as the authors note there are reasons why this ought to have been true in the past. So, while Scenario 4 makes up a reasonable xtra assumption (the one I describe in my post), the authors' favored scenario does not explain easily how a species that inhabited Western Eurasia (Neandertals) ended up contributing not-too different amounts of DNA to Europeans and Chinese. So, for the time being, I'm sticking to my guns and saying that the paper has uncovered something important, but probably not Neandertal admixture.

Science 7 May 2010:
Vol. 328. no. 5979, pp. 710 - 722

A Draft Sequence of the Neandertal Genome

Richard E. Green et al.

Neandertals, the closest evolutionary relatives of present-day humans, lived in large parts of Europe and western Asia before disappearing 30,000 years ago. We present a draft sequence of the Neandertal genome composed of more than 4 billion nucleotides from three individuals. Comparisons of the Neandertal genome to the genomes of five present-day humans from different parts of the world identify a number of genomic regions that may have been affected by positive selection in ancestral modern humans, including genes involved in metabolism and in cognitive and skeletal development. We show that Neandertals shared more genetic variants with present-day humans in Eurasia than with present-day humans in sub-Saharan Africa, suggesting that gene flow from Neandertals into the ancestors of non-Africans occurred before the divergence of Eurasian groups from each other.

Link

69 comments:

eurologist said...

The authors do have an explanation of why any admixture occurring in Europe left little trace: by that time, AMH population sizes were already large, overwhelming any additional genetic inflow, and no longer hugely expanding (in fact, much of European population expansion may have occurred via continued immigration - without additional admixture).

That is, admixture that happen shortly out-of-Africa affected the entire subsequent and expanding population and "froze in" - admixture events in Europe on the other hand became (almost) meaningless.

I like the common-cnacestor divergence age (~800,000 years), which makes sense in the context of Heidelbergensis. I am still leery about the 270,000 to 440,000 gene flow date, which sound too young to me - unless early Neanderthal populations back-entered Africa in large numbers. I have previously discussed such an idea that AMHs may be a derivative of such a mixed population, originally situated in North Africa. But are the segments shared with Eurasians younger than 300,000 - 400,000 years?

I agree that this comparison should be done with a much larger number of African genomes to shed light on this. One would have to carefully correlated with also-tested percentage of European admixture in such populations.

Maju said...

I think Dienekes that you are proposing model 4, which the authors do not consider disproven, though they favor model 3 because of parsimony reasons.

Model 4 would be the existence of a structured population in Africa prior to the OoA with the would-be-migrant branch already more akin to Neanderthals already.

I think model 4 is quite unlikely and unnecessarily complicated. It would require that the OoA population would have gone extinct in Africa later on or otherwise we'd detect those genes for instance among Yoruba and not among San, which represent the oldest separate branch without doubt.

It is also indirect evidence that the admixture happened in West Asia prior to arrival and expansion in South Asia, East Asia and Oceania. Which is interesting as we should be talking of the Abbassia Pluvial, when there is evidence of interactions among both species at Palestine, with adoption of Mousterian and certain penetration in Egypt.

Or maybe there is the adoption by the main migrant group of an admixed remnant of a previous migration. Who knows?

One answer, so many questions.

vasa said...

Seem to me little bit unfair to assume that there were just Neanderthal and modern sapiens at the time of out-of-Africa spreading. What about other populations? How were they related to those mentioned groups? Probably closely. I think we are not able to distinguish admixtures of Neanderthal from admixtures of Neanderthals-or-human related populations. What I want to say is that modern sapiens in Asia could mixed with some descendant of Jinniushan man (that can be somewhere between Neanderthal and modern man) and now it seems he mixed with Neanderthal. And I am pretty sure there were plenty of such different populations in that times.

Dienekes said...

I think Dienekes that you are proposing model 4, which the authors do not consider disproven, though they favor model 3 because of parsimony reasons.

On the contrary, their model fails the parsimony test since East Asians end up with similar "Neanderthal" admixture as Europeans, without explaining how this is possible.

So, while my hypothesis (their Scenario 4) posits the existence of population structure in Africa, their hypothesis must resort to some geographical acrobatics to explain how "Neanderthal" admixture is evenly spread out in Eurasia.

There is population structure in Africa today, with at least a dozen distinct clusters, one of which is related to West Eurasians. I'd say it is rather the notion that African Homo sapiens at the time of OOA was panmictic that requires a special explanation.

Dienekes said...

The authors do have an explanation of why any admixture occurring in Europe left little trace: by that time, AMH population sizes were already large, overwhelming any additional genetic inflow, and no longer hugely expanding (in fact, much of European population expansion may have occurred via continued immigration - without additional admixture).

That would explain why late European Neandertal admixture into AMH would be low, but it's a double-edged sword. Demographic dominance of AMH over Neandertals would also result in substantial introgression of European AMH genes into Neandertals, but the authors find "Thus, all or almost all of the gene flow detected was from Neandertals into modern humans."

Similarly, we can explain the low introgression of Amerindian genes into the white American gene pool by the numbers' advantage that the latter enjoyed, but at the same time, this numbers' advantage led to substantial introgression of European genes into Amerindians.

Marnie said...

eurologist,

I wasn't implying that you had made any comments in previous threads about Neanderthals.

But I do believe you've made some comments about a distinct paleolithic contribution in Germany.

So I'm not sure if you are "happy" with these results.

The Svante Paabo group has been working on this for years. It's nice to finally see the paper.

Now I have to worry about how much Neanderthal I have in me and what that means. Well, it's a good distraction from everything else that's going on in the world.

Marnie said...
This comment has been removed by the author.
Marnie said...

"I was wary of this paper's conclusion as soon as I realized that the authors contended that Europeans and East Asians did not differ significantly in their levels of Neanderthal admixture. You see, Neandertals were absent from East Asia, so there is no reason for East Asians to have such admixture at all, or to have much less of it than West Eurasians do.


"Even if we assumed that undifferentiated Eurasians picked up Neandertal admixture in West Asia, the fact would remain that Europeans co-existed with Neandertals after the ancestors of East Asians had moved on towards the Pacific. So, they ought to have more Neandertal admixture -if such admixture ever took place. Indeed, one of the main arguments for introgression of Neandertal genes in modern humans is the long co-existence of modern humans with Neandertals in Europe."

Dienekes,

Come on! The authors are saying that they believe that most of the Neanderthal admixture happened 100,000 in SW Asia, shortly after the out-of-Africa departure, and before the European-Asia departure.
I think that's plausible, especially since the population size of people who left Africa can be guessed to have been low.

In any case, I'm sure the authors will now set to work on diffentiating the Asia and Northern European Neanderthal components. So I'll just wait for the next paper from these guys.

Maju said...

Dienekes: you are suggesting structure in Africa prior to the OoA (as it's highly doubtful that there were any other Homo spp. in that continent after the OoA, when the L0, L1, L2 and L3 expansions had already taken place. By suggesting that the ancestors of modern Africans mixed with other unspecified and unknown Homo spp. in Africa more than the ancestor of modern Eurasians but still in Africa, you are proposing a variant of model 4, the only difference being that you emphasize hypothetical admixture with other unclear populations, while the authors only consider Neanderthal admixture.

I think both scenarios are essentially the same and fail the parsimony test (what does not disprove them, just make them less easy to believe).

You say that there is a problem with East Asians, Papuans and West Eurasians having exactly the same Neanderthal ancestry but that is not a problem if admixture happened right at the OoA process and not afterwards. And your "alternative" does not change anything about that anyhow.

Between Qahfez and Jawalpuram there are some 40,000 years, if not more. The period of Qahfez is also the one when Neanderthal-Sapiens contact was most intense. Even if those early Palestinians, who very likely are at the origin of the now discovered admixture process, did not star the OoA themselves it's very possible that they interacted with other Homo sapiens communities of the Red Sea or elsewhere in West Asia, transferring some Neanderthal genes in the process (usually introgression happens via an intermediate group, not directly between the two main groups).

This is model 3, which is extremely parsimonious.

Maju said...

"... but the authors find "Thus, all or almost all of the gene flow detected was from Neandertals into modern humans."

"Similarly, we can explain the low introgression of Amerindian genes into the white American gene pool by the numbers' advantage that the latter enjoyed, but at the same time, this numbers' advantage led to substantial introgression of European genes into Amerindians".

But that is because we are talking of two different species. Remember that we were considering until yesterday the no admixture scenario as something quite likely.

Also of all three Vindija specimens, only one is more recent than 40 Ka. And the other specimens to be published are also older than these dates. Hence it's not likely that they would have incorporated Sapiens genes. A different scenario might have existed where Neanderthals adopted UP tech (Chatelperronian-Szletian) but for that we would have to analyze those populations specifically.

For instance the famous Neanderthal woman that was reconstructed not so long ago and whose "photo" illustrates so many articles on Neanderthals was a likely hybrid specimen and in her we may well find Sapiens genes not found in Vindija.

andrew said...

"What about Africans? Why do they stand further away from Neandertals? The answer is simple: low-level of admixture with archaic humans in Africa itself."

Just to flesh out John Hawks answer to this question, his conclusion is that the methodology of the study would not reveal any Neanderthal roots shared by both African and non-Africans, because genes shares by all modern humans and by Neanderthals look like genes of a shared common ancestor statistically.

I'd also point out that if there was demic replacement in Europe from West Asian populations in the Neolithic, that any impact of admixture in the paleolithic between Neanderthals and paleolithic Europeans would be greatly muted, quite possibility to the point where the impact of this admixture was smaller than the 1-4% confidence interval range of Neanderthal contribution could distinguish.

If 2% of our DNA on average has a Neanderthal source, and 80% of current European DNA has a Neolithic West Asian source (as suggested by Y-DNA and mtDNA estimates that have been published), then a Neanderthal contribution anywhere in the range from 0% to 12% in paleolithic Europeans would be essentially indistinguishable.

We know from ancient DNA that paleolithic modern human Europeans differed significantly from the existing populations there (by more than the different between modern Europeans and aboriginal Australians) and that there were phenotypic differences as well. So a "Neolithic eraser" effect is plausible.

West Asian populations should have the lowest percentage of Neanderthal contributions in the European-West Asian sphere (and hence the closest DNA to Asian DNA), because Neaderthals were probably replaced there before they were replaced elsewhere.

Dienekes said...

But that is because we are talking of two different species. Remember that we were considering until yesterday the no admixture scenario as something quite likely.

I don't see how that is relevant, as the authors contend that the two species were not reproductively isolated. One would need an additional assumption, such as Neandertal-AMH offspring incorporated only in Neandertal societies, but then you are adding assumptions to the "parsimonious" model.

onur said...

Assuming that the new Neanderthal genetic introgression scenario is true (though it has its own problems as demonstrated by Dieneke), the important question to be asked is what lasting physical and mental effects this introgression produced among the Neanderthal-admixed anatomically modern humans (also what lasting advantages or disadvantages it conferred). I very much doubt such effects are uniform among the Eurasian modern humans even if their Neanderthal admixture is evenly distributed. From the other side of the coin, I also wonder the practical lasting consequences of the lack or relative paucity of the Neanderthal genes among the sub-Saharan modern humans.

Andrew Oh-Willeke said...

"what lasting physical and mental effects this introgression produced among the Neanderthal-admixed anatomically modern humans (also what lasting advantages or disadvantages it conferred). I very much doubt such effects are uniform among the Eurasian modern humans even if their Neanderthal admixture is evenly distributed."

The question is a bit narrower than that, I think. The question is what advantages or disadvantages it conferred in the time and place that the Neanderthal genetic introgression took place and became more or less uniformly distributed in a presumably West Asian or East African subgroup gene pool.

Once Neanderthal genes become embedded more or less uniformly in an expanding modern human gene pool, even if the Neanderthal genes confer little benefit, so long as there is not strong selective pressure against those genes, the effects are likely to be pretty uniform among modern Eurasian modern humans. This is because any selective disadvantages that come from the Neanderthal genes will be outweighed by the overall selective advantages of the cultural and genetic package that they are attached to.

If the effective population size of the Out of Africa population is 2000 (which is on the order of some of the estimates made from other genetic data) then a 1%-4% introgression of Neanderthal genes implies an effective introgressive Neaderthal population of 20-80 Neanderthals. If the overlap is on the order of 1000 years in any one place (as suggested here), and the Eurasian similarities derive from just one overlap, then there could b just a few introgressive Neanderthals per generation if the modern human population were growing only slowly at the start of the Out of Africa period, and it would take only a thousand years or two of that West Asian source population for all Eurasian modern humans to be bottled up in a single community for those introgressions to be fully dispersed in the gene pool.

If the conditions of hunter-gatherer life 40,000 years ago were pretty similar to those that the Neanderthals had adapted to after leaving Africa hundreds of thousands of years earlier, then almost all Neanderthal distinctive traits (presumably better than or neutral with respect to traits where there was no genetric change because they were a better fit to non-African environments) would have, if not a strong adaptive value, at least not strongly negative adaptive value.

For example, perhaps Neanderthals had an ammino acid that better progressed small quantitites of mammoth fur in the digestive system (a trait useless in Africa but helpful in Neanderthal Europe), and also had a trait that dealt better with metabolism in a lower winter temperature environment. The metabolism trait might be important enough to bring a package of Neanderthal genes into the genome (even if it was eventually selected away from in some or all modern human populations, something suggested by the lack of Neanderthal mtDNA in modern humans), while the ammino acid trait might linger unaltered in all subsequent populations since it conferred neither an advantage or disadvantage.

Marnie said...

"the important question to be asked is what lasting physical and mental effects this introgression produced among the Neanderthal-admixed anatomically modern humans."

Onur,

They're all crazy of course.

onix said...

dieneke said:
What about Africans? Why do they stand further away from Neandertals? The answer is simple: low-level of admixture with archaic humans in Africa itself.

altho in or two other points of your post i recognise the same or a similar thought,this examplifies. i strongly disagree.

firstly amh are a result of a mixture of archaic hominids per say. secondly i think it is obvious that with the hominids largely evolving in africa, the circumstances there were most conductive to their evolution (wich is basically the option for gene flow through populations(1)(to express)).

i have one more remark i haven't posted before, i think it is well possible that the genetical evidence represents admixture of several eurasian subspecies in the pool. in that case the suggestion is the genetical diversity and difference between those was not very big. at least not any bigger then between africans and europeans, that might be a result of a similar archaic gene flow as i assume inside africa, and it may be interesting to test when that becomes possible.

on a sidenote i wonder if really the nr of 'out of african's' was so small, this research implies (even explicitly) so, but another possibility appears that preferential selection was for the similar genetic expressions over the whole line. wich is not at all contradicted with the research.

a reason to doubt the very small nr of emigrants is actually mdna, with a total of at most 40 odd mdna's. (probs closer to 30) the option is a whole 7.5 % of the african population migrated.

(1)in animals you would say sub-species.

Micke Böghög said...

So was the possible Neanderthal contribution entirely absent in the african individuals or was it present although much less than in Eurasians?

eurologist said...

We know from ancient DNA that paleolithic modern human Europeans differed significantly from the existing populations there (by more than the different between modern Europeans and aboriginal Australians)

Andrew, could you provide me with a reference to that (you mean autosomal, I assume)?

West Asian populations should have the lowest percentage of Neanderthal contributions in the European-West Asian sphere (and hence the closest DNA to Asian DNA), because Neaderthals were probably replaced there before they were replaced elsewhere.


Not in the model(s) considered by the authors, in which the Neanderthal contribution entered right before or at OOA.

I agree with Marnie, I am sure one of the next studies will try to see if there is a difference in the Neanderthal elements between various Eurasian humans.

On the advantages, at the least the Neanderthal element helped widen the gene pool of the few AMHs that made it OOA. I also agree that there would be diet, metabolic, and temperature related advantages - some of which may not have come to play out until many 10,000 years later, when AMHs entered more northern latitudes.

Also, at least northern Neanderthal populations were stuck at close proximity for many months at a time (in the winter). So there may be some differences in terms of cooperation, mediating fights, or even mental tasks/ group activities that made the long winter periods bearable (perhaps superficially cultural, but with some genetic component).

So was the possible Neanderthal contribution entirely absent in the african individuals or was it present although much less than in Eurasians?

Again, they cannot tell with their current approach. Any Neanderthal component that is present in all current human populations simply appears as old, unchanged throughout. However, it is clear that the particular contribution that made it into Eurasians is not present in Africans.

German Dziebel said...

Neanderthals also share shovel-shaped incisors with Asians and Amerindians, an important ancient hominid dental feature which is rare in modern Sub-Saharan Africans.

terryt said...

"a reason to doubt the very small nr of emigrants is actually mdna, with a total of at most 40 odd mdna's".

Mitochondrial DNA provides the best evidence in favour of very small numbers involved in the OoA. Just two of the several contemporary L3 lines made it out, and L3 itself is just one of many mtDNA lines that existed in Africa at the time.

"at the least the Neanderthal element helped widen the gene pool of the few AMHs that made it OOA".

Very true, and something I've maintained as being necessary for a long time.

"Remember that we were considering until yesterday the no admixture scenario as something quite likely".

Some of us have never been entirely convinced of 'the no admixture scenario' though.

"the methodology of the study would not reveal any Neanderthal roots shared by both African and non-Africans, because genes shares by all modern humans and by Neanderthals look like genes of a shared common ancestor statistically".

And we can be fairly sure that migration back into Africa has occurred at times, so the admixture level may be greater than what the present research suggests.

eurologist said...

And we can be fairly sure that migration back into Africa has occurred at times, so the admixture level may be greater than what the present research suggests.

The author's gene flow date (versus the divergence date) suggests as much. However, assuming a much larger population of humans in Africa than emerging Nenderthals in Europe/ West Asia, and assuming only a small fraction of those back-migrating with the Saharan Pump, how much gene material total, and how much "new" could they have contributed that wasn't already within the much larger African gene pool? Especially considering the relatively short initial separation time.

Once you go towards 60,000 - 100,000 years ago (~400,000 to 800,000 years of separation), and then consider the impact on a few thousand AMHs, it's a completely different story.

Maju said...

"I don't see how that is relevant"...

What I mean is that the low admixture levels cannot be explained only in intra-species (i.e. H. sapiens) demographic terms but in a significantly less permeable inter-species interaction. Otherwise, H. sapiens would have been just absorbed into the Neanderthal gene pool at its arrival from Africa and there would have been no OoA as we know it.

I understand that H. neanderthalensis, as the native species with similar technology had the demographic advantage in Eurasia early on. So that explains why there was so little flow.

Also, in a separate but related digression, Sapiens introgression in West Asian Neanderthals may have never arrived to Europe, even if it was bidirectional. West Asian Neanderthals may have never backmigrated to Europe at all.

But whatever because admittedly I don't have all the details really clear at this stage.

Marcel F. Williams said...

Obviously, if there was any admixture of modern humans with Neandertals, it occurred in the areas where Neandertals use to exist (Europe and the Middle East).

The fact that Asian populations appear to have just as many Neandertal genes as modern European populations may suggest that there has been substantial gene flow between Europe and Asia over the past 35,000 years. The fact that Neandertal genes appear to be substantially less evident in sub-Saharan populations may suggest that gene flow between sub-Saharan Africa and Eurasia is substantially less than gene flow between Europe and Asia.

During the glacial periods, sub-Saharan Africa was probably almost completely isolated from Eurasia thanks to the expanse of Sahara desert during the dry periods. And even during the warm interglacial periods, gene flow into Eurasia is severely restricted to the meager land bridge corridor out of Africa via Egypt while geographic contact between Europe and Asia is much more extensive.

Maju said...

"I very much doubt such effects are uniform among the Eurasian modern humans even if their Neanderthal admixture is evenly distributed".

Has anyone got clear by now if the 1-4% admixture is fixated for all Eurasians or is just like any other normal gene: variable?

...

"a reason to doubt the very small nr of emigrants is actually mdna, with a total of at most 40 odd mdna's. (probs closer to 30) the option is a whole 7.5 % of the african population migrated".

Why do you say that. The OoA population surviving haploid lineages are only 2-3 Y-DNA ones and 2 mtDNA ones. Even within L3 alone (a fraction of a fraction of a fraction of a fraction of a fraction of a fraction of a fraction of African mtDNA as a whole) the Eurasian-specific lineages M and N are only 2/7 of known basal branches (excluding whatever L3* is left), that is 28.5% only.

I get only a 0.4% stake of the African mtDNA diversity. Though the exact population involved could be larger or smaller in proportion to the whole of the species in Africa, it gives at least a reference, more than an order of magnitude smaller than yours.

So no idea where you got that figure from.

"(1)in animals you would say sub-species".

No because subspecies implies a certain marked genetic differentiation significantly larger than the one found between any two human populations.

"We know from ancient DNA that paleolithic modern human Europeans differed significantly from the existing populations there (by more than the different between modern Europeans and aboriginal Australians)"

That's simply not true.

And there's only mtDNA data from those ages in humans so far.

onix said...

thx terryt and maju, i am not a great adherer to genetical research, and i cant even pretend to know more then the crudest of the compilation of it all.

it was my understanding (apparently wrongly) that where we have 2 different mdna's in eurasia (with presumably a couple more in the most metropolic states and city's (london netherlands rome eg.)) there are over 30 still replicating (..) in africa.

apparently these numbers are incorrect. i said it because i considered that a substantial part of the population of say egypt and nubia might have moved north without bringing in more then that nr of mdna's, since i also have read the rest of the mdna's is largely represented under the equator.

apparently i was wrong, and apparently there are (still?) more, i consider that good news.
thx.

onix said...

about the subspecies thing i am not so sure however. we call birds with minimal difference in looks a subspecies when,in general, they have a different geographical location.

i am quite sure we would call say white foxes that have white fox baby's and black foxes that have black ones subspecies. for me it doesn't matter as a racial thing, since i never even considered (eg.) neanderthalers substandard.
(another example 'd be i think of heidelbergensis as neanderthaler and/or (evolved) erectus)

what made me raise the point is actually the new australopithecus (sub)species. personally i assume archaic hominids had a more 'subspecies like' distribution then more modern ones. altho time must have been a limiting factor.

it seems of interest when we consider how geneflow took place inside africa in archaic hominids (and quite possibly homo)and that mechanisms outside africa would adhere to a similar model.

simply put i don't expect morphological differences translate in equally obvious genetical ones. (in either archaic hominids or amh)

look for example at orca subspecies, altho they can be rather different in size and looks, and completely different in their way of life, as far as i know there is (just like in humans) no genetical barreer whatsoever to stop them from interbreeding. so i think subspecies shouldn't be a laden term. it means just : the same species with a different look.
is that the point?

i think after all that especially in archaic hominids the term will still be very handy, to understand their evolution, looks and distribution.

German Dziebel said...

@Terry: "Mitochondrial DNA provides the best evidence in favour of very small numbers involved in the OoA. Just two of the several contemporary L3 lines made it out, and L3 itself is just one of many mtDNA lines that existed in Africa at the time."

I think evidence, mtDNA or other, points to the greater rate of lineage loss outside of Africa due to selection and/or demography. The fact that populations outside of Africa are closer to Neanderthals is the most recent indicator that this is the likeliest scenario to explain the greater diversity of modern African populations. It's complete nonsense to attribute the greater genetic proximity of non-Africans to Neanderthals to some kind of quick boost of genes from Near Eastern Neanderthals at the exit from Africa. Science doesn't work by attributing a systematic phenomenon (at least genetic and odontological on the paleobiological side) to a historical accident. The only reason the authors prefer scenario 1 is because they feel obligated to salvage the out of Africa hype.

What in fact happened is that a small population branched off Neanderthals or Asian Homo erectus, speciated somewhere in the Americas (on the strength of linguistic and kinship evidence, in the very least) and then migrated back into the Old World, replaced all hominids and peopled Africa. The rate of mutation wasn't constant or may be some older alleles were under negative selection and newer alleles were under positive selection, so we ended up with having an illusion that humans originated in Africa. The existence of the Mungo man sequence is a further proof that there used to be more past diversity in Eurasian humans than there's diversity in modern African humans.

Dienekes said...

So was the possible Neanderthal contribution entirely absent in the african individuals or was it present although much less than in Eurasians?

It depends on whether you allow back-migration into Africa. If yes, then it would be present, if no, no.

Dienekes said...

Come on! The authors are saying that they believe that most of the Neanderthal admixture happened 100,000 in SW Asia, shortly after the out-of-Africa departure, and before the European-Asia departure.

Neandertals lived on for another 70,000 years or so, so why did the admixture "stop". There is absolutely no evidence, by the way, that East Asians are descended from Out-of-Africans who left 100,000 years ago.

You say that there is a problem with East Asians, Papuans and West Eurasians having exactly the same Neanderthal ancestry but that is not a problem if admixture happened right at the OoA process and not afterwards. And your "alternative" does not change anything about that anyhow.

I don't see how it is more parsimonious to say that the admixture happened "right at the OOA". AMH and Neandertals co-existed after the OoA process in West Eurasia, so you'll have to come up with a reason why they didn't keep on mixin'

Another problem with the authors' scenario is that it proposes zero or close to zero modern-to-Neandertal admixture, which makes little sense as AMH were the dominant species and it's difficult to see why admixture would not -at least- be both ways.

Marnie said...

"Neandertals lived on for another 70,000 years or so, so why did the admixture "stop"."

"andrew's" comments that the admixture may not have stopped, but is buried in the uncertainty of the result, are instructive.

"There is absolutely no evidence, by the way, that East Asians are descended from Out-of-Africans who left 100,000 years ago."

We shall likely soon have more of an answer to that long standing question.

Maju said...

Onix:

"it means just : the same species with a different look. is that the point?"

Subspecies does not have any agreed meaning: any meaningful subdivision you make under the species level would be that. For example there could be a blond subspecies and even a Muslim subspecies...

But if we get serious and want to use some standards of certain biological value throughout species, we have to compare the genetic distances, not just appearance, location and behaviour and do so with relation to the standards used among other species than humans, and, when we do so, the human intercontinental genetic differences are just too low: the species is too young and has not diverged enough to form anything we could really call subspecies in ways that it's standard for other mammals.

At most we can talk of "variants" which is not standard nomenclature but it's used anyhow, specially in gardening.

Maju said...

Dienekes:

"Neandertals lived on for another 70,000 years or so, so why did the admixture "stop""...

Why do you think?

I understand that this admixture is product of a "brief" episode (many millennia) in Palestine/West Asia. When H. sapiens moved eastward to South Asia, the admixture, which was anyhow rare, as the evidence shows, stopped altogether.

Alternatively there were several migrant groups only one of which really had such intimate contact with Neanderthals, being fused before the arrival to South Asia.

Whatever the case, it's apparent that the Eurasian H. sapiens population did not interbred with Neanderthals anymore, not even when they back-migrate to West Eurasia later on.

The story is in fact talking about a rather strong inter-species barrier (not necessarily biological but at least behavioral and of mating preferences) that was only occasionally crossed when the migrant population was still small enough to leave trace.

"There is absolutely no evidence, by the way, that East Asians are descended from Out-of-Africans who left 100,000 years ago".

Uh? There's lot of evidence: Y-DNA and mtDNA conspire for a single common origin in Africa. You can argue about the exact date of the migration but not really about the single shared origin.

"AMH and Neandertals co-existed after the OoA process in West Eurasia, so you'll have to come up with a reason why they didn't keep on mixin'".

I am surprised that they ever mixed at all, so lack of admixture is just normal for me. What I'd like to know is why they did mix in the first encounter. It may be that prejudices about each other had not yet grow big enough, it may be that there was a small population only which interbred, later transfering some of that package to other migrant H. sapiens (a classical case in introgression examples) but I'm not so surprised they did not interbred anymore in any case. What surprises me is that they actually did, so keeping it circumscribed to certain episode early on, when it's known that H. sapiens even adopted locally Neanderthal tech (Mousterian) makes good sense. Later that would just not happen (though it's arguable that Neanderthals did adopt "Sapiens tech" in some cases).

There is a gap anyhow between Palestine 120,000 BP and Palestine 58,000 BP. Maybe we had just culturally established by then that Neanderthals were "monsters" or maybe even "food" and not just "people".

The cultural attitude of the different ethnic populations may totally affect the outcome. Anyhow, there was a strong, though obviously permeable, inter-species barrier limiting the flow all the time. Otherwise, Neanderthals would just have absorbed the migrants and we would not exist as such.

But that did not happen because our ancestors, still mostly Sapiens, found a place to thrive in South Asia and beyond before total assimilation happened in West Asia.

terryt said...

From the paper:

"gene flow from Neandertals into the ancestors of non-Africans occurred before the divergence of Eurasian groups from each other".

So the mixture happened very soon after the OoA. To paraphrase someone on another blog, does that mean that most here believe that Neanderthals lived just across the Bab al Mandab, in Yemen?

"The existence of the Mungo man sequence is a further proof that there used to be more past diversity in Eurasian humans than there's diversity in modern African humans".

To me it indicates that the current post should be taken at face value: human groups mixed as they moved around (always assuming that the different Mungo Man mtDNA is correct). Haplogroup lines are continually becoming extinct. The interesting thing is that the human haplogroups tend to be much less diverse, with a more recent common ancestor, than those of other species. To me this indicates that some sort of selection other than genetic is at work in humans. Cultural? Technological?

"how much gene material total, and how much 'new' could they have contributed that wasn't already within the much larger African gene pool? Especially considering the relatively short initial separation time".

The 'relatively short initial separation time' is not the relevant factor in this case. The 'new' genetic material introduced would be a product of hybridisation, so the relevant separation time is that between modern and Neanderthal humans: at least half a million years. Some Neanderthal genes were almost certainly not 'already within the much larger African gene pool'.

"we call birds with minimal difference in looks a subspecies when,in general, they have a different geographical location".

Along with most mammal species, and even invertebrates. And I agree that humans vary with 'geographical location'. So, if not the same thing as 'subspecies', they are certainly a product of the same process that leads to subspecies in other species.

"personally i assume archaic hominids had a more 'subspecies like' distribution then more modern ones".

Almost certainly so, a fact usually ignored when considering which particular 'subspecies' gave rise to our ancestors. It was probably a combination of a number of such subspecies.

onur said...

But if we get serious and want to use some standards of certain biological value throughout species, we have to compare the genetic distances, not just appearance, location and behaviour and do so with relation to the standards used among other species than humans, and, when we do so, the human intercontinental genetic differences are just too low: the species is too young and has not diverged enough to form anything we could really call subspecies in ways that it's standard for other mammals.

At most we can talk of "variants" which is not standard nomenclature but it's used anyhow, specially in gardening.


Maju, from what I understand, neither for all species nor only groups of species (such as mammals) nor even single species there is any fixed genetic or other kind of threshold of subspeciation. Every single species is evaluated independently and the genetic clusters (or, to use your words, variants) in the highest levels within a species are considered its subspecies. Current race categories (the archetypal ones) in humans more or less fit this description as shown by countless genetic tests.

catperson said...

Dienekes makes an interesting point/ Perhaps there was no breeding between humans and neandertals; instead the reason why non-African DNA resembles neanderthal DNA is that the region of Africa that gave rise to neandertals was also the same region that gave rise to non-African humans. Which is logical because that region was probably close to the geographic exits out of Africa so it's not surprising that the same gene pool would leave twice (once as incipient neandertals and much later as non-African humans)

Maju said...

"does that mean that most here believe that Neanderthals lived just across the Bab al Mandab, in Yemen?"

See here.

Re. LM3 ("Mungo man"), there's been a lot of justified criticism to that old and surely not repeatable test. However I am recalling now the almost forgotten B006 X-DNA lineage that certain studies considered as apparently too old and/or distinct and hence suspicious of differential origin. Does anyone keep a link?

Marnie said...

Onur,

We've had this discussion about human variants and sub-species many times and we have all been informed, more than once, about your thoughts on race, sub-race and sub-species.

I, myself, could never imagine discussing face-to-face with my neighbour across the street, or my neighbour two doors down, or in fact with you, the fact that you are probably a different "sub-race" or "sub-species" from me.

Except for the most obvious differences in appearance, which we don't really need a classification system for, I can't account for what we would accomplish from a sub-race or sub-species classification system.

I'm fine with talking about meaningful populations, clusters and variants. Words do matter.

For example, in terms of classifying Africans and African Americans, for medical purposes, the race or sub-race classification system utterly fails. Why? Because the genetic makeup of Africans and African Americans is too complex to lend much value to easy medical diagnosis based on race (skin color and facial features). Even classification based on yDNA or mtDNA would certainly fail.

Sure, there are a few diseases that do seem to fall along racial lines, but most do not. Most of our efforts to diagnose along racial lines have been failures.

Race, as perceived, is not of much medical use. Most of the great advances in medicine have been public health advances, or advances that work universally, in any case. (Clean water, antibiotics, elimination of smallpox, vaccination.)

No one is special because of their self perceived race, not even Gioiello. Of course, Gioiello would disagree as he thinks he is an Italian God. I seem to have met him before although I'm sure I have not.

I myself would be very suspicious of someone who clings strongly to a need to classify people based on skin color and facial features. Maybe we should start to put people who do that in a sub-race category: the weak self-esteem, need to obsessively classify sub-race, inflated ego category.

Let's get the geneticists working on finding that gene, right away. I'm sure we could form a sub-race out of all the people who have that gene and therefore a compulsive need to classify people based on a narrow set of observable "looks". (and subsequently elevate themselves.)

I was going to suggest putting them on an island, but with global warming, all the islands are disappearing. Well, there should be some space opening up in the Canadian North sometime soon. We could put them up there, so long as the wolves and the Inuit agree.

Well, I'm sorry for such a long diatribe about how much these sub-race discussions annoy me. It must be my Neanderthal alter ego.

Maju said...

Chewing on the details of the paper (actually taking a pseudo-break from doing that as I write) I notice two curiosities in them:

1. Table 2 and related section of the article mention specific alleles (78 SNPs in c.70 genes) that are fixated in the derived form among us and are instead ancestral among Neanderthals, i.e. mutations that define what is being H. sapiens.

Several related to skin (sweat glands for example, melanin), hair, smell, vision, sperm and nuclear regulation ("specifity"), etc. Complicated to evaluate but interesting in any case.

Also they mention HARs (regions with clear evolution in the human lineage specifically): most are shared with Neanderthals (i.e. happened before divergence) but a handful (1.6% or so) are specific of H. sapiens.

2. I think I perceive a small but consistent pattern in the Japanese sample (and specially in 2 individuals: 50%) to be "less Neanderthal" than the rest and particularly the Han. Chinese might appear as very very slightly "more Neanderthal" instead. The samples are small and the differences quite low and maybe non-significant... but they are consistent across the two chips and various measures.

It is intriguing at least.

Maju said...

PS- This last point is about table 4 data.

princenuadha said...

Can someone explain how an ancient genetic substructure within africa could have arisen so that the ancestors of modern non-africans were more related to Neanderthals than the ancestors of modern Africans.

one assumption that must be made is that at the time of AMH contact with European Neanderthals, AMH in Europe (and Asia) were significantly more related to AMH in Africa.

by the above assumption pre-eurasians must have mostly split from pre-africans after mostly splitting with pre-european Neanderthals.

the only scenarios I can envision are:
1) at least two Neanderthal populations; one which had gene flow with pre-eurasians while the pre-european Neanderthals continued to differential from the pre-eurasians. This would explain why there was little to no gene flow with Neanderthals and AMH in Europe.

2) There was basically one Neanderthal population and after the pre-eurasian and pre-african split the pre-eurasians had limited gene flow with the Neanderthals early on... then stopped. The gene flow stopped for long enough so that the pre-eurasian differentiation from the Neanderthals could great large enough differences that when early Europeans re-encountered Neanderthals there was insignificant additional Neanderthal admixture to the European gene pool.

Maju said...

"Can someone explain how an ancient genetic substructure within africa could have arisen so that the ancestors of modern non-africans were more related to Neanderthals than the ancestors of modern Africans".

Not me. It's just an alternative hypothetical scenario not really supported by any known data. It would require (at the current state of knowledge) that the ancestral Eurasian population would have vanished or nearly so after the OoA, what is extremely inconsistent with the known facts about mtDNA L3 and Y-DNA E, both of which must have been part of that proto-Eurasian population and which experienced expansion (not contraction, much less extinction) in Africa at the OoA timeline and later.

I think we can forget about this hypothesis, really.

...

Addendum to my previous post: after revision of the data there are some inexactitudes in point 2 but there is still some minor but apparent excess Neanderthal input among Han (n=2). The Japanese individual (n=1) appears to have the opposite trend but not just it's a too small sample but the comparisons with different Yorubas give extremely opposite trends (low Neanderthal admixture and high Neanderthal admixture), so I'd rather discard this case unless further evidence comes forward.

But the Chinese "slight excess" seems very consistent and is maybe significant.

eurologist said...

The 'relatively short initial separation time' is not the relevant factor in this case. The 'new' genetic material introduced would be a product of hybridisation, so the relevant separation time is that between modern and Neanderthal humans: at least half a million years. Some Neanderthal genes were almost certainly not 'already within the much larger African gene pool'.

Terry, I don't think you understood what I was talking about. Basically, it is reasonable to assume that very early Neanderthals (or late Heidelbergensis) would have been able to back-migrate during the Saharan pump events. And the difference of the gene flow date (400,000 years) to the divergence date (800,000 years, matching early Heidelbergensis) suggests as much.

My argument was that this type of African substructure was too early, and that at that time much or the gene flow material would have been part of the much larger African pool, anyway. Of course there would have been also another 300,000 years of mixing in Africa to wash it all out. Conversely, at around 100,000 years ago, the two populations had diverged more, and now you are also talking about just a tiny subset of the African pool that was affected: just the few people that made it OOA (perhaps in two known steps, like Maju posted on his site).

Dienekes said...

so the relevant separation time is that between modern and Neanderthal humans: at least half a million years.

Actually the authors estimate this as 272-435 thousand years; and since they use West Africans as representatives of modern humans, if they used Eurasians (who are closer to Neandertals) the separation time would be even younger.

Maju said...

@Dienekes:

That's only because of the parameters of the MC equations they have chosen to use.

But which is the archaeological justification for such a low age guesstimate?

None. There's some tentative anthropometric studies but nothing, absolutely nothing, in the cultural/technological aspect.

So the situation is:

H. ergaster > H. heidlbergensis > Neanderthal

H. ergaster > H. rhodesiensis > H. sapiens

With a material splinter at the Acheulean difussion period c. 800 Ka - 1 Ma ago.

We can have long "coffee discussions" on MC and anthropometric hunches but they are evidence of nothing unless supported by something else.

We lack that 'something else'.

onur said...

A new article from John Hawks, which opposes the argument of no interbreeding between sapiens and neanderthals:

http://johnhawks.net/weblog/reviews/neandertals/neandertal_dna/african-population-structure-neandertal-mixture-2010.html

onur said...

I myself would be very suspicious of someone who clings strongly to a need to classify people based on skin color and facial features. Maybe we should start to put people who do that in a sub-race category: the weak self-esteem, need to obsessively classify sub-race, inflated ego category.

Let's get the geneticists working on finding that gene, right away. I'm sure we could form a sub-race out of all the people who have that gene and therefore a compulsive need to classify people based on a narrow set of observable "looks". (and subsequently elevate themselves.)


Those who know me know that I have no interest in any ego or nos thing. In fact, I have a strong tendency to think independent of myself and my personal and social connections (including biological ones). Racial classification of people is not a simple matter of looks, it goes deep into biology, hence genetics. Genetic study after study after study after study after study after study after study after study after... have confirmed the reality of biological human races, also confirming the results of anthropometric (including craniometry) studies, which have been being made since much earlier than the age of genetic studies.

Dienekes said...

So the situation is:

H. ergaster > H. heidlbergensis > Neanderthal

H. ergaster > H. rhodesiensis > H. sapiens


Not true, Middle Pleistocene African and European hominids belonged to a single taxon that gave birth to both Neandertals and Homo sapiens.

Katerina Harvati, Petralona: Link Between Africa and Europe? Chapter 2 in New directions in the Skeletal Biology of Greece (eds. L.A. Schepartz, S. C. Fox, C. Bourbou) 2009

What you are describing is the accretion hypothesis which is explicitly rejected in the cited book chapter.

A new article from John Hawks, which opposes the argument of no interbreeding between sapiens and neanderthals:

http://johnhawks.net/weblog/reviews/neandertals/neandertal_dna/african-population-structure-neandertal-mixture-2010.html


What he observes is easily explained by admixture by anatomically modern humans into Neandertals -if it needs explanation at all. He also takes at face value the Neandertal-modern split time, which is an overestimate under Scenario 4, given that West Africans are used as a proxy for modern humans to estimate that split.

Note that the authors can only really reject modern admixture into Neandertals, if they make the assumption that the non-Africans that admixed with them were already closer to Yoruba than to San. However, there is no clear reason to think that for early anatomically modern sapiens.

In conclusion, the recent coalescence dates can be easily explained by admixture from moderns into Neandertals before or shortly after the Yoruba-San split.

Maju said...

"Not true, Middle Pleistocene African and European hominids belonged to a single taxon that gave birth to both Neandertals and Homo sapiens".

That's arguable, not demonstrated. Skeletal biology and molecular clock are not rocket science, you know.

And the lack of any such migratory track in the archaeological record after Acheulean makes me suspect that it's wrong.

German Dziebel said...

"Re. LM3 ("Mungo man"), there's been a lot of justified criticism to that old and surely not repeatable test. However I am recalling now the almost forgotten B006 X-DNA lineage that certain studies considered as apparently too old and/or distinct and hence suspicious of differential origin. Does anyone keep a link?"

Look up Zietkiewicz et al. Haplotypes in the dystrophin DNA segment point to a mosaic origin of modern human diversity // Am J Hum Genet. 2003 Nov;73(5):994-1015. Earlier on Dienekes: http://dienekes.blogspot.com/2009/05/x-chromosomes-and-settling-of-americas.html.

B006 is the basal lineage and it's found at high frequencies in Asia and the Americas and at low frequencies in Africa. Same biogeographic pattern as the shovel-shaped incisors shared by modern humans with Neanderthals. The same pattern is further displayed by the "nuclear insert" (Zischler et al. 1995. A nuclear ‘fossil’ of the mitochondrial D-loop and the origin of modern humans. Nature 378, 489–492) to which the Mungo Man sequence in question was the closest.

catperson said...

I assume the scientists who did this study ruled out the possibility that the genetic similarity between non-African humans and neanderthals was not just the result of convergent evolution (i.e. both populations evolving to the Eurasian environment rather than mixing). If there really was mixing it seems strange that it has not been detected in mitochondrial DNA as usually the dominant population (if it mixes) mates with the females of the conquered population.

waggg said...

@ catperson :

I've read a comment about it elsewhere :

http://evoandproud.blogspot.com/2010/05/ask-rhino.html?showComment=1273189583416#c5633721397132786148

"As for alleles meshing well in such a closely related species: happens often. Taurine and zebu cattle are the result of separate domestications, from stocks of wild cattle that diverged around 500k years ago - more divergent than human and Neanderthals. But zebu genes have introgressing like crazy into an originally taurine African stock - because they work better in the heat and aridity. Creeping zebuization has affected the Middle East as well over the past few thousand years.

On the other hand, they don't mesh every time, which is undoubtedly why we don't see any Neanderthal mtDNA or Y-chromsome lineages today. This indicates that Neanderthal mtDNA had a selective disadvantage of, say, half a percent. Plausible if they were energy wasters.

By the way, we see cattle populations in Egypt that are a quarter zebu (on autosomal genes) but don't show any zebu mtDNA _or_ y-chromosomes. "

eurologist said...

I assume the scientists who did this study ruled out the possibility that the genetic similarity between non-African humans and neanderthals was not just the result of convergent evolution (i.e. both populations evolving to the Eurasian environment rather than mixing).

Developing the same exact mutation for a given phenotype is the extreme exception, not the rule. There are just too many genes involved, and too many possible mutations to reach a similar goal (usually simply enhancing or diminishing some gene expression). Just look at the many mutations for skin, hair, and eye color - and those are quite variable for quick adaptation. Yet, there are tens of ways to do the same (or similar)trick. In less variable genes, or in circumstances where many genes contribute in a very complicated manner, it would be yet much less likely to happen.

Thus, parallel evolution rarely brings about the same mutation.


If there really was mixing it seems strange that it has not been detected in mitochondrial DNA as usually the dominant population (if it mixes) mates with the females of the conquered population.

But which population was dominant? The answer is both: Neanderthals were initially more populous, AMHs - at least at one point - seemed to be more flexible and more adaptable through technological (rather than bodily) changes.

I also like the dial-step process that Maju is thinking about, just because it can explain very easily why only AMH mt-DNA and Y-DNA survived: the conventional OOA group (~60,000 to 70,000 years ago) was about 50,000 years more modern than the original small band in Palestine, and may have been successful all on their own. Or perhaps, the first migration left no trace, and the second one simply took on and merged with small bands of Neanderthals along the way to India, long before encountering larger Neanderthal groups, say in Turkey or the Balkans or central Europe.

catperson said...

Developing the same exact mutation for a given phenotype is the extreme exception, not the rule. There are just too many genes involved, and too many possible mutations to reach a similar goal

I don't know much about genetics, but this argument sounds counterintuitive to me. Is it really so rare for the same mutations to occur independently? Think of all the diseases and syndromes we see that are caused by some random mutation that many unrelated people have the misfortune of being independently afflicted by.

And while there are multiple ways of reaching the same goal, some may have a better chance of occuring than others.

Further some traits are fairly simple and thus it may be probable that these simple mutations could occur independently in different populations. Keep in mind that both populations were in a stressful cold environment & stress is now known to trigger mutations.

It's also possible that the mutations occured in Africa before the ancestors of humans and neanderthals diverged, however the mutations only came under high selection when each population left Africa and needed these mutations to survive the colder climate. Meanwhile, the mutations had no value in Africa and gradually vanished in Africans, thus causing Africans to be genetically less similar to neanderthals than non-Africans are.

I just think the results of this study can be interpreted in a number of different ways and people are just jumping to the breeding conclusion because it's interesting and provocative. But we already have strong (though not conclusive) mitochondrial evidence that breeding did not occur, and the two populations were isolated for so long that makes us think they could breed even if they wanted to?

Maju said...

"I don't know much about genetics, but this argument sounds counterintuitive to me. Is it really so rare for the same mutations to occur independently?"

It is counterintuitive to me that mutations occur in parallel, with exactly the same changes (because of what Eurologist explained: they are random, not pre-selected).

However I read recently on certain research on butterflies that indicated that in more than one different species the same mutation had developed independently for the same function. I'm talking from memory but I think they explained this because of functional constrictions because of previously shared evolutionary paths.

Anyhow, considering that we don't know at this moment if the Neanderthal alleles in our species are adaptative or merely neutral and that all the rest (or almost) of the Eurasian genome is made up of the pre-existent African diversity, and considering that the distance between Eurasians and Africans is virtually the same for all cases but that is not what happens with the overall genome (where some Africans are clearly closer to Eurasians than others), my opinion is that the Neanderthal origin is almost unquestionable... unless it is H. ergaster/erectus introgression instead.

This last possibility is also perfectly viable, specially if we accept that Neanderthals and us diverged some 900 Ka ago. and not just that ridiculous figure of 300 Ka. (please, we would not have become so radically different is such a short time!) from some variant of H. ergaster/erectus that was carrier of Acheulean technology. If so, what we see as "Neanderthal" genes would rather be general Eurasian Homo spp. genes, distinct from the African ones that predominate in our species.

This is also possible. But whatever the case, it is almost certain that Eurasian hominins other than H. sapiens are at the origin of that 2% of genes. Otherwise Yorubas would have most of them too, as they are much closer to Eurasians than Bushmen.

"Keep in mind that both populations were in a stressful cold environment & stress is now known to trigger mutations".

Eurasians migrated by the tropical corridor along the south of Asia. It was never really cold probably yet. Adaptation to cold, essentially techno-cultural, happened upon the scatter of Eurasians, not at the OoA migration - not yet.

Not sure why do you say that "stress triggers mutations". I can conceive that stress emphasizes selection, destroying those less apt maybe, but why would it cause mutations as such?

"It's also possible that the mutations occured in Africa before the ancestors of humans and neanderthals diverged"...

That can be totally discarded, I understand. If so we'd see them in Africa in differential form, with Yorubas (high in Y-DNA E and mtDNA L3, i.e. closely related to Eurasians) having them at least in some apportions and Bushmen not. This simply does not happen, so your scenario is practically impossible.

"Meanwhile, the mutations had no value in Africa"...

And why would they have no value in Africa and have it in New Guinea? Beats me!

Anyhow, they are surely neutral alleles, fixated by random drift in all or most cases.

Maju said...

"But we already have strong (though not conclusive) mitochondrial evidence that breeding did not occur"...

Nope. The mtDNA data was no evidence of interbreeding but could never totally rule out such possibility, just made it less likely and surely not large.

What Green et al. have found now is not large, so it fits perfectly with the mtDNA evidence. Minor admixture would not normally leave any evidence in the haploid lineage pool because these tend to get fixated in the majority clades, not the odd ones, by mere statistical logic. But, by the same statistical logic, all ancestors, Neanderthal included, have their proportional chance to leave an overall genetic mark according to their numerical relevance at the original events (OoA period in this case).

So most likely 1-4% of Neanderthal-specific alleles means 1-4% (roughly) of Neanderthal ancestry.

"... and the two populations were isolated for so long that makes us think they could breed even if they wanted to?"

A lot of closely related species can interbreed. Even some belonging to different genus, such as lion and tiger. However this never (or almost never) happens in the wild (but polar and grizzly bears do interbreed occasionally in North America naturally).

Neanderthals and Sapiens were either two closely related different species or two very distinct different subspecies of the same species. This is arguable but the discussion on the biological classification should not obscure the fact that they were probably inter-fertile but surely with some barriers also, making such interbreeding difficult, instinctively undesirable and sometimes even not working at all.

Even being inter-fertile, being also so different morphologically and surely in customs and psychology, would be a barrier to full assimilation of one species with the other, making it practically difficult. In other words: would you let your daughter marry a N Neanderthal? Would she find him attractive? Same for guys. The answer is most probably no.

But there's always someone with odd tastes, you know. And we can't exclude rape as well.

And if life along Neanderthals was peaceful and productive for generations, affections and aesthetic preferences, which are largely cultural, learned, should become more favorable, also allowing for intimate situations to happen more frequently.

Whatever the case, here it is. The last legacy of Neanderthals.

eurologist said...

This last possibility is also perfectly viable, specially if we accept that Neanderthals and us diverged some 900 Ka ago. and not just that ridiculous figure of 300 Ka. (please, we would not have become so radically different is such a short time!) from some variant of H. ergaster/erectus that was carrier of Acheulean technology. If so, what we see as "Neanderthal" genes would rather be general Eurasian Homo spp. genes, distinct from the African ones that predominate in our species.

Of course, we have to be a bit careful here. Are we talking about old genes that got preserved in Eurasia but not in Africa, or new adaptations/changes in Eurasia?

I think the second option is much more likely, since I would maintain that there was a much larger gene pool in Africa, actually preserving more ancient genes than Eurasia could (unless perhaps we go back to erectus, which would also necessitate Neanderthal-erectus interbreeding to shuffle those genes first into Neanderthal, and then again ending up with likely just too little material left to transfer to AMHs).

So, IMO the ~2% the authors see is largely stuff that first surfaced and accumulated in Eurasia over the past ~million years, so the short name "Neanderthal" is not all that misleading if we all accept them to derive from Heidelbergensis in significant isolation from Africa.

In other words: would you let your daughter marry a N Neanderthal? Would she find him attractive? Same for guys. The answer is most probably no.

Hehe... I always find it sooo seductive to day-dream (i.e., write imaginary screen plays) about possible scenarios.

Let's not forget the wallflower one. You know, there is that truly helpful, polite, yet quiet Neanderthal boy that got adopted after he showed the thirsty clan the much-needed fresh water source. Given your somewhat ugly and simply too smart daughter that no one wants to take even as second or third wife, perhaps not a bad choice? What father could resist? ;)

Maju said...

"Are we talking about old genes that got preserved in Eurasia but not in Africa, or new adaptations/changes in Eurasia?"

If I'm correct, what the authors have measured is alleles that were mutated in Neanderthals and not in all humans with relation to the Pan-Homo common ancestor.

Are you saying this is mere adaptative noise? What's in Yemen that is not in Eritrea? What's in Palestine that is not in Egypt? What's in Indonesia that is not in Congo?

I fail to see the need for adaptation in the journey from the Nile to the Jordan. And I fail to see how European Neanderthals would have been better adapted to those southern Asian climates than AMHs who lived in North Africa and Ethiopia since some 190 Ka ago (attested by the archaeological record).

"So, IMO the ~2% the authors see is largely stuff that first surfaced and accumulated in Eurasia over the past ~million years, so the short name "Neanderthal" is not all that misleading if we all accept them to derive from Heidelbergensis in significant isolation from Africa".

It's not. After all Neanderthals are the ones providing the reference - and most likely admixture happened with them anyhow, because we know that there were contacts in the Middle East 100-plus Ka ago.

German Dziebel said...

It's worthwhile looking at the book Reticulate Evolution by Arnold, in which all the reported cases of "introgression" between "archaics" and modern humans are cataloged and analyzed. Available on Google books.

What's interesting about these studies is that the highest frequencies of those "introgressed" alleles are often reported from American Indians. I'm sure if American Indians were in Green's sample, they would have been by a tad closer to Neanderthals than the Chinese. For instance, haplogroup B of human lice DNA, which is found mostly in North and South America and which diverged from more frequent haplotype A about 1.2 MM years ago, is claimed by Reed et al. 2004 to have been introduced into modern humans through a "reproductive contact" with Asian Homo erectus. Later the same haplogroup B was detected in Europeans and Australian aborigines (these are our Eurasian humans from Green et al. 2010) but not in Africa. See Light et al. 2008; Raoult et al. 2008.

princenuadha said...

"Now I have to worry about how much Neanderthal I have in me and what that means"

hahahaha, I had the same reaction... its just a natural thought. seeing how it's basically just a Eurasian thing I'm not concerned.

catperson said...

Not sure why do you say that "stress triggers mutations". I can conceive that stress emphasizes selection, destroying those less apt maybe, but why would it cause mutations as such?

Wasn't there a major revision of evolutionary theory a few years back that found that mutations are not simply random but rather they are more probable in organisms that are stressed. Evolutions way of increasing the odds that an organism that is having trouble adapting will have a better chance of mutating into a form better suited to its environment. It was a big deal at the time because it suggested that Lamarkism might not have been totally wrong.

catperson said...

That can be totally discarded, I understand. If so we'd see them in Africa in differential form, with Yorubas (high in Y-DNA E and mtDNA L3, i.e. closely related to Eurasians) having them at least in some apportions and Bushmen not. This simply does not happen, so your scenario is practically impossible.

But my argument is that perhaps the mutations entered the human line right before the ancestors of neanderthals left Africa hundreds of thousands of years ago (and preserved in neanderthals who needed them out of Africa), thus would have gradually decreased in Africa because they had little or no selective advantage there. By the time modern humans first left Africa about 60 thousand years ago, they may have decreased in Africa to only 1-4% and preserved at this level in non-African modern human populations because of their selective advantage in non-African environments. Meanwhile they continued to decrease (eventually to 0%) in Africans where they continued to be useless.

And why would they have no value in Africa and have it in New Guinea? Beats me!

Perhaps the ancestors of those who live in New Guinea had some exposure to colder climates as they moved to New Guinea. Some probably came from East Asia.

Maju said...

"But my argument is that perhaps the mutations entered the human line right before the ancestors of neanderthals left Africa hundreds of thousands of years ago"

That's model 4: pre-existent structure in Africa that the authors say the can't disprove but seems less likely.

I understand that the population from which the OoA group(s) splintered is best described by mtDNA haplogroup L3. This lineage expaned in Africa at that same time and is very important in all African peoples except a handful of isolated hunter-gatherer surviors, such a the Bushmen.

That means that if there was structure in Africa prior to the OoA in that sense, Yorubas must have anywhere between 25-100% the Nenaderthal component Eurasians show. And that is not the case at all.

So this admixture episode did not happen in Africa, at least not south of the Sahara (there's a chance that the episodes of Palestine permeated to Egypt, where some Mousterian is also found AFAIK).

catperson said...

I understand that the population from which the OoA group(s) splintered is best described by mtDNA haplogroup L3. This lineage expaned in Africa at that same time and is very important in all African peoples except a handful of isolated hunter-gatherer surviors, such a the Bushmen.

That means that if there was structure in Africa prior to the OoA in that sense, Yorubas must have anywhere between 25-100% the Nenaderthal component Eurasians show.


I don't think so because Yorubas stayed in Africa so they would have gradually lost the mutations in the last 60,000 years if it was only being selected outside Africa.

My argument is that these mutations first occured in the population that is ancestral to both neanderthals and modern humans but since they were only useful outside Africa, they gradually decreased the longer folks stayed in Africa. Since neanderthals were the first descendants of the ancestral population to leave Africa, they preserved the mutations at the highest frequency. Since modern human Eurasians left Africa much later, they only carried small amount of the mutations (1 to 4%) and since Yorubas & bushmen never left Africa, they completely lost the mutation.

So this admixture episode did not happen in Africa, at least not south of the Sahara

My argument is that is NOT an admixture episode since I'm arguing that it occured before the ancestors of modern humans and neanderthals diverged into separate groups. Mixing implies two different populations came together, but I'm arguing they were the same population when the mutations emerged.

Maju said...

"... they would have gradually lost the mutations in the last 60,000 years if it was only being selected outside Africa".

That would demand a selective mechanism that you have failed to describe. Climatically and so on, PNG is much more like Nigeria than either is like Europe.

"My argument is that is NOT an admixture episode since I'm arguing that it occured before the ancestors of modern humans and neanderthals diverged into separate groups".

There's a long AFRICAN history for the line leading to H. sapiens alone after such thing happened. Longer than all the time passed since the OoA, even in the most recentist scenarios.

It's not logical at all.

catperson said...

That would demand a selective mechanism that you have failed to describe. Climatically and so on, PNG is much more like Nigeria than either is like Europe.

Perhaps the mutations were selected in colder climates and perhaps the population that migrated to Papua New Guinea experienced colder climates prior to arriving there.

There's a long AFRICAN history for the line leading to H. sapiens alone after such thing happened.

The long African history explains why there was enough time for the mutations to decline to only 1 to 4% in the population that became Eurasians. Perhaps one can look at these mutations as a clock measuing the time when a group left Africa. Neanderthals hundreds of thousands of years ago, so they had lots of the mutations, Eurasians only 60 K ago so they had only 1 to 4%, and Africans never, so they have 0%. But I'm speculating with a limited grasp of these issues so I apologize for any logical errors.

Maju said...

"Perhaps the mutations were selected in colder climates and perhaps the population that migrated to Papua New Guinea experienced colder climates prior to arriving there".

By the well attested coastal (tropical) route?!

Perhaps... not.

We are talking of alleles that are in the derived state in Neanderthals in comparison with modern humans and chimpanzees. It's not an archaism but an innovation: a Neanderthal-specific innovation.

Thorstein Veblen said...

This might be a stupid question, but with how much certainty do we know that Neanderthals were confined to Europe? That seems strange to me, given that the rest of Eurasia has a similar climate and ecosystem. If they could find a niche in Europe, why couldn't they find a niche in Central and East Asia as well? We also know that historically, most Paleontologists have come from Europe...

And, to the extent that Neanderthals were not confined to Europe, then it's not strange that we find their genes roughly equally present in all non-African DNA...

Maju said...

Thorstein: As of now it's known that Neanderthals inhabited in West Asia and Central Asia, in addition to Europe. There's no trace of them in East or South Asia, nor of their favorite technology: Mousterian.

foreign body said...

how do you explain the absence of RH negatives in Africa? Could the rhesus negative be related to Neanderthals? I am RH neg and i know that i could not have children with an RH+ man without modern drugs. RH neg men on the other hand can have a progeny with RH+ women. Maybe Neanderthals had the same problem when they encountered Homo Sapiens (RH+) and they were decimated as a result.
My DNA shows no African matches whatsoever...but Siberian, Native American matches, besides Aryan and Celtic matches.