June 13, 2014

An older layer of Eurasian admixture in Eastern Africa

The authors propose that a genetic component found in Horn of Africa populations back-migrated to Africa from Eurasia ~23 thousand years ago. I did not read this carefully yet, but it seems reasonably plausible. The migration uncovered by Pagani et al. and Pickrell et al. (~3kya) is probably not the whole story of Eurasian back-migration into Africa; that episode probably involved only Semitic speakers (it's hard to imagine any other language being carried from the Middle East at that time frame). But the Eurasian affiliation of East Africans extends well beyond Semitic speakers. In North Africa this is even clearer as the native (pre-Arab) population is definitely broadly West Eurasian and this must have come about by back-migration.

The back-migration of M1 and U6 into Africa seems to be at a similar time depth as the one proposed by these authors (post-initial UP, but pre-Neolithic). I have proposed that haplogroup E may represent an even earlier layer of Eurasian back-migration, while the signals identified by Pagani et al. and Pickrell et al. in East Africa, Pickrell et al. for Southern Africa, and the limited (but present) Neandertal ancestry in Yoruba and Pygmies documented by Pruefer et al. represent later events. The arrival of Arabs and Europeans are later events still.

For a time, there was a taboo against imagining back-migration into Africa; in a sense this was reasonable on parsimony grounds: Africans have most autosomal genetic diversity and the basal clades of mtDNA and Y-chromosomes; a model with Out-of-Africa is simpler than one with both Out-of and Into-Africa. However, we now know that pretty much all Africans have Eurasian ancestry, ranging from at least traces in theYoruba and Pygmies (to account for the Neandertal admixture) to intermediate values in East Africans, to quite  a lot in North Africans.

Eurasian admixture in Africa seems to be general, variable, and to have occurred at different time scales. It's still the best hypothesis that modern humans originated in Africa initially and migrated into Eurasia. However, it is no longer clear that Africa was always the pump and never the destination of human migrations.

PLoS Genet 10(6): e1004393. doi:10.1371/journal.pgen.1004393

Early Back-to-Africa Migration into the Horn of Africa

Jason A. Hodgson

Genetic studies have identified substantial non-African admixture in the Horn of Africa (HOA). In the most recent genomic studies, this non-African ancestry has been attributed to admixture with Middle Eastern populations during the last few thousand years. However, mitochondrial and Y chromosome data are suggestive of earlier episodes of admixture. To investigate this further, we generated new genome-wide SNP data for a Yemeni population sample and merged these new data with published genome-wide genetic data from the HOA and a broad selection of surrounding populations. We used multidimensional scaling and ADMIXTURE methods in an exploratory data analysis to develop hypotheses on admixture and population structure in HOA populations. These analyses suggested that there might be distinct, differentiated African and non-African ancestries in the HOA. After partitioning the SNP data into African and non-African origin chromosome segments, we found support for a distinct African (Ethiopic) ancestry and a distinct non-African (Ethio-Somali) ancestry in HOA populations. The African Ethiopic ancestry is tightly restricted to HOA populations and likely represents an autochthonous HOA population. The non-African ancestry in the HOA, which is primarily attributed to a novel Ethio-Somali inferred ancestry component, is significantly differentiated from all neighboring non-African ancestries in North Africa, the Levant, and Arabia. The Ethio-Somali ancestry is found in all admixed HOA ethnic groups, shows little inter-individual variance within these ethnic groups, is estimated to have diverged from all other non-African ancestries by at least 23 ka, and does not carry the unique Arabian lactase persistence allele that arose about 4 ka. Taking into account published mitochondrial, Y chromosome, paleoclimate, and archaeological data, we find that the time of the Ethio-Somali back-to-Africa migration is most likely pre-agricultural.

Link

48 comments:

terryt said...

"For a time, there was a taboo against imagining back-migration into Africa; in a sense this was reasonable on parsimony grounds: Africans have most autosomal genetic diversity and the basal clades of mtDNA and Y-chromosomes; a model with Out-of-Africa is simpler than one with both Out-of and Into-Africa".

As I just wrote at your post on Y-DNA K, surely it is now impossible for anyone to believe that 'modern humans' arose in a single region and expanded around the world from that region. Backwards and forwards, to and fro, it's always been happening.

Vincent said...

The time proposed by the author is compatible to a back-migration of E-M215 to the Horn of Africa. E-M215* is possibly 25-30 ky old and the only example we know is an Iranian from Khorasan. I can't say for sure that E-M215 originated there, but certainly it arose somewhere in West Eurasia. It migrated into Africa possibly through the Levant (or through South Arabia, given the presence of the sub-clade E-M281 there). There are no other plausible candidates for this admixture: Ethiopian J-M267 is virtually al J-P58 (therefore much younger than 20 ky) and T-M70 is T1a1a-L208, younger than 20 ky and found with low diversity in the Horn.

astenb said...

@ Vincent. Again you are going to have to read older available studies on SUB SAHARAN AFRICANS to know that E1b1* ancestral is found in the Sahel, West Africa as well as the Horn of Africa. It is quite frequently found in nearly every study that analyzes Ethiopian Y-dna. The precursor to E-M215 is already present. Furthermore You have the V38/100 + M329 split from E1b1* Horn Africans, particularly Southern Ethiopian and Omotic speakers are dominated by E-M215 derivatives as well as the M329 lineages. These lineages are found in populations with low or lacking "Ethio-Somali" genetic component : Ari Blacksmiths - See Plaster et al. And take note, Omotic speakers...in this study the Ari are supposed to have the Autochthonous presence prior to "Ethio-Somali".

Looking at these dates E-M78 with a Northern Origin and E-M293 with a Southern one are already estimated at 21KYA... the date of the supposed admixture event. OTOH M35 is older in Africans with a lowbound of 31kya. Already too old to have an origin in the Levant and represent this Admixture event. The parent clade Pn2* Also found in the horn?... 38kya. [Hirbo et al]

Tying anything to the backmigrtion of E subclades based on CURRENT evidence is pretty much a complete instant failure. Your ideas of Backmigrating E clades are nonsensical. I would suggest taking a look at the PN2* page on wikipedia and getting familiar with the sources and its distribution in Africa.

Fiend of 9 worlds said...

That's what archaeology has always said. Riffians skulls are practically the same as northern europeans.

Anonymous said...

How does Hofmeyr fit into this? Or is it just too early to count (although the dating could be wrong)?

Vincent said...

"Again you are going to have to read older available studies on SUB SAHARAN AFRICANS to know that E1b1* ancestral is found in the Sahel, West Africa as well as the Horn of Africa."

There is no more E-P2* since Trombetta et al. (2011). You say that in your comment, "older studies" found it , because they couldn't test M281 and V38. We only have one E-M215* from Eastern Iran, E-M2* in West Africa and E-M329 in Ethiopia/Yemen. E-P2* originated in the Levant, together with E1b2-P75, which (as I have shown in my comment to the U6 article) has only been found in two English, one Palestinian and one Canary Islander (possibly of European descent).

"Tying anything to the backmigrtion of E subclades based on CURRENT evidence is pretty much a complete instant failure. Your ideas of Backmigrating E clades are nonsensical."

I think that "my ideas" are not so crazy on the basis of current evidence.

"I would suggest taking a look at the PN2* page on wikipedia and getting familiar with the sources and its distribution in Africa."

Thanks for the suggestion, I go now and take a look. Hmm, none of the old studies cited in the Wikipedia article has tested for M281 or V38 SNPs, and I think that if these two SNPs were tested for, none of the samples would be E-P2*.

terryt said...

"There is no more E-P2* since Trombetta et al. (2011)".

E-P2(xV38) is listed in the latest ISOGG update:

http://www.isogg.org/tree/ISOGG_HapgrpE.html

Vincent said...

terryt,

You misunderstood what I wrote. I meant, virtually all African cases which were reported as E-P2* in old studies have been resolved by the new V38 and M281 SNPs. Presumably, the only two real cases of E-P2* (could be V38, but I find it extremely unlikely) were found in 0.38% of 523 Anatolians (Abu-Amero et al. 2009). The same study reports real E-M96* at 1.27% in Saudi Arabia and 0.55% in Lebanon. Zalloua et al. (2008) report E(xE1,E2) in Syria. Even if we exclude the Bulgarian E-M96*(xM35, M2) and the Albanian E*(xM35), real cases of E-M96* are only found in Eurasia, because Ethiopian E-M96*xM35,M2,M329) found by Plaster (2011) is likely M281+, and none of the E-M96* reported cases in sub-Saharan Africa has been tested for V38.
In conclusion, E-M96*, E-P2* and E-P75 are certainly Eurasian.

GailT said...

[i]"it is now impossible for anyone to believe that 'modern humans' arose in a single region and expanded around the world from that region. Backwards and forwards, to and fro, it's always been happening."[/i]

You use evidence of a back migration 23 kya to make claims about hypothetical migrations that [i]might[/i] have occurred 200 kya. You seem to be fixated on the old Wolpoff multiregionalism theory. The genetic evidence still suggests that migrations 200 kya had minor contributions to admixture between Eurasia and Africa. I look forward to any new evidence that might challenge that consensus, but as D. notes, [i]"It's still the best hypothesis that modern humans originated in Africa initially and migrated into Eurasia."[/i].


German Dziebel said...

@gailT

""It's still the best hypothesis that modern humans originated in Africa initially and migrated into Eurasia."[/i].

It's a total eclipse of reason: the "best" hypothesis is the one that has no facts to support it. We have "archaic" Eurasian (Neandertal and Denisovan) alleles in Africa but no "archaic" African alleles outside of Africa. The best scientific hypothesis is that a modern Eurasian population carrying Y-DNA DE lineages entered Sub-Saharan Africa and admixed with local hominin populations carrying hgs A00, A and B. Those lineages appear outside of Africa only after 1492.

Fiend of 9 worlds said...

Every other animal had a multiregional origin. Humans are supposed to be special for some reason. Right.

The reason this (which as I said is long since known) mostly kills the idea of out of africa (which has no archaeological support of any kind) is that in the actual out of africa papers the migrations OU of africa are not much further back than these migrations IN to africa! So they have to be viewed as debunked theories. I have viewed then that way for some time now.

That doesn't mean every possible out of africa scenario is debunked but there are other archaeological reasons that go against other time periods as well. Basically some small group of genetic people try to make an argument that goes completely against the body of archaeology. Sorry but one is wrong, and it's obvious who it is - the ones who don't actually know anything about archaeology!

We also see this same kind of y-dna and mtdna replacement happen in other species like lions, which we absolutely know didn't come out of one place and completely replace existing populations of lions from elsewhere. This coincidence of one cladal gene having one group (which probably has to come from asia if the recent paper on the split of K and Q and R is true anyway) really means absolutely nothing.

terryt said...

"You use evidence of a back migration 23 kya to make claims about hypothetical migrations that [i]might[/i] have occurred 200 kya."

I agree that it's still the best hypothesis that modern humans originated in Africa initially and migrated into Eurasia'. But it is very unlikely that those humans trace their ancestry back to some Homo erectus population that had remained in Africa since they first emerged from some Australopithecus species. In fact we don't know the history of surviving haplogroups beyond some 200,000 years ago. In fact there is some degree of evidence that Homo erectus populations moved between Africa and Eurasia. It is also extremely unlikely that human ancestors emerged from Africa no more than twice over the whole two million years of their existence and moved into Africa only once.

"Presumably, the only two real cases of E-P2* (could be V38, but I find it extremely unlikely)"

Perhaps, but surely it is quite likely they are E-M215, many branches of which have been found in Eurasia.

"Zalloua et al. (2008) report E(xE1,E2) in Syria".

That is interesting. How many individuals?

"In conclusion, E-M96*, E-P2* and E-P75 are certainly Eurasian".

We're going round in circles here. We need to wait for more research.

terryt said...

"You use evidence of a back migration 23 kya to make claims about hypothetical migrations that [i]might[/i] have occurred 200 kya."

Actually not only that migration. We now know that Y-DNA R and Q's deeper ancestry is in SE Asia. So we have IJK starting off in SW Asia, LT dropping off KLT somewhere in South Asia, NO dropping off somewhere in SE Asia and P originating in Southern Wallacea. From where that haplogroup moved back to eventually become the dominant Y-DNA through the whole of the northern half of Eurasia. With that movement alone we have evidence for considerable backwards and forwards movement.

Vincent said...

"Perhaps, but surely it is quite likely they are E-M215, many branches of which have been found in Eurasia."

You may be right. I checked and saw Cinnioglu et al. (2004) did not test for M35. Therefore the two haplotypes are most likely E-M35*, even if some of their STR values are quite unusual when compared to E-M35* haplotypes I have studied.

"That is interesting. How many individuals?"

There are 2 out of 286 individuals in Syria (but there's possibly a third, reported as simply E) and 5 out of 916 individuals in Lebanon (again one other individual is reported as simply E). From the same study there is also 1 individual out of 107 who is reported as simply E in Akka, Israel along with 2 out of 164 Cypriots.

GailT said...

"Humans are supposed to be special for some reason. Right."

It should be obvious that modern humans are special. Using our intelligence and technology, we've expanded throughout out the world and even travelled to the bottom of the ocean and to the moon. If that's not special, I don't know what is.

GailT said...

But it is very unlikely that those humans trace their ancestry back to some Homo erectus population that had remained in Africa since they first emerged from some Australopithecus species. In fact we don't know the history of surviving haplogroups beyond some 200,000 years ago. In fact there is some degree of evidence that Homo erectus populations moved between Africa and Eurasia. It is also extremely unlikely that human ancestors emerged from Africa no more than twice over the whole two million years of their existence and moved into Africa only once.

There are no surviving y-DNA or mtDNA haplogroups beyond about 200 kya, so the uniparental DNA presents compelling evidence for a recent origin of AMH in Africa. But I agree the picture might be more complex, and hopefully autosomal DNA will fill in the complexity. It seems likely that were multiple waves of expansion out of Africa whenever climate conditions allowed. The question is whether any of those populations survived and whether there is evidence for back migrations into Africa.

I can imagine scenarios in which archaic humans did or did not migrate back to Africa. If you are suggesting that AMH could have arisen during 200 to 100 kya as a result back migrations during that period, that seems very speculative without DNA or physical evidence, but perhaps evidence will be discovered with improved analysis of autosomal DNA and more testing of ancient remains.

But going back to the original comment, the fact that there is evidence for back migrations 23 kya does not make it impossible to believe that AMH might have evolved in Africa 200 kya without admixture with Eurasian populations. Other scenarios are possible but they are speculative until there is data to support them.

terryt said...

"Every other animal had a multiregional origin. Humans are supposed to be special for some reason. Right".

Correct. I have this idea that the idea humans descend from a single small population, unlike other species, is as a result of centuries of biblical ideas of racial distinctiveness and the Garden of Eden and Noah's ark beliefs of single origin. Basic beliefs are difficult to discard.

"We also see this same kind of y-dna and mtdna replacement happen in other species like lions, which we absolutely know didn't come out of one place and completely replace existing populations of lions from elsewhere".

We aslo have an interesting situation in the case of European and American bison where the two 'species' don't share mt-DNA. European bison mt-DNA is much closer to that of cattle. And American mallards don't share mt-DNA with Eurasian mallards. American mallard DNA is related to that of other related American dabbling ducks.

"This coincidence of one cladal gene having one group (which probably has to come from asia if the recent paper on the split of K and Q and R is true anyway) really means absolutely nothing".

I think there is no doubt that the surviving lines of both mt- and Y-DNA have their origin in Africa. But haploid DNA is just part of the story. And we don't know the deeper origin of either haploid DNA.

"The best scientific hypothesis is that a modern Eurasian population carrying Y-DNA DE lineages entered Sub-Saharan Africa and admixed with local hominin populations carrying hgs A00, A and B".

For once I agree with German. It is far more likely that Y-DNA DE entered Africa than that Y-DNA E did. However I presume German still believes DE originated in America which is actually impossible to accept.

"Those lineages appear outside of Africa only after 1492".

Yes, but the Y-DNA phylogeny tells us that DE formed from a single branch within B and that the ancestors of both A and B branched off early from the CT line that led to DE.

Unknown said...

I don't know if I'm reading this chart right (Table 4 on page 11) but how is it that the Ethio-Somali component is divergent from the Ethiopic component only by eight thousand more years? And Nilo-Saharan and Niger-Congo components roughly the same time? All these languages do have Y- Hg E in common too. Perhaps E did originate in the Arabian peninsula 55,000 years ago but then migrated to East Africa again 34,000 years ago when the Niger-Saharan-Ethiopic component diverged first and then 23,000 years ago the Ethio-Somali component followed after? Sounds out there but what else could explain their closeness? Any ideas?

terryt said...

"It seems likely that were multiple waves of expansion out of Africa whenever climate conditions allowed. The question is whether any of those populations survived and whether there is evidence for back migrations into Africa".

We know genes from two of those migrations survive today: Neanderthals and the mysterious 'Denisovans'. The reason others are yet to be noted is probably that we don't yet know what we should be looking for.

"It should be obvious that modern humans are special. Using our intelligence and technology, we've expanded throughout out the world and even travelled to the bottom of the ocean and to the moon. If that's not special, I don't know what is".

All the 'special' aspects you mention are cultural, not biological. Biologically we are not special.

"There are no surviving y-DNA or mtDNA haplogroups beyond about 200 kya, so the uniparental DNA presents compelling evidence for a recent origin of AMH in Africa".

Your comment, 'Using our intelligence and technology, we've expanded throughout out the world' explains the very limited diversity of human haplogroups as compared to other species. Cultural and technological advances have allowed particular human groups to dominate whole regions.

@ Vincent:

Thanks for that clarification.

Rokus said...

"Such a genetic distance makes three thousand years for a haplogroup rapidly expanding from SE Asia to Central Asia (P-P27: 27,200 ka => MA-1, 24,000 ka) quite impossible".

'The Romani, or Gypsies, made it all the way from northern India to the far west of Europe in much less than 3000 years. Something like 400 years'

The genetic distance of those few Gypsy lineages that may still be derived from their Indian origin never caught our attention for being as disproportionate.

'I think there is no doubt that the surviving lines of both mt- and Y-DNA have their origin in Africa.'

There is, because I think otherwise. Successive migration waves could have left relict DNA in a conservative place like Africa. Changes may have radiated from somewhere else. Both YDNA and mtDNA are subject to natural selection, including for the human species that apparently experienced several important gender-related and metabolism-related changes within a short period of time. Moreover, the current phylogeny is a human construct that more than coincidentally serves the contemporary mindset. As long we don't have ancient DNA from Africa there should always be doubt, for many reasons.

Fiend of 9 worlds said...

y-dna and mtdna is supposed to mean nothing about race, yet it's supposed to mean everything about the entire origin of all living people. It's obviously a belief that comes entirely out of political bias, and Paabo has had to work hard to justify the fact we see more and more neanderthal dna around the world including all through africa as well. If there was some justice, it would be recognized he was wrong and he'd be kicked off the faculty and someone who had predicted admixture would replace him.

GailT said...

@terry - I don't understand how anyone could agree with the ridiculous claim that: "The best scientific hypothesis is that a modern Eurasian population carrying Y-DNA DE lineages entered Sub-Saharan Africa and admixed with local hominin populations carrying hgs A00, A and B".

By characterizing y-DNA haplogroups A and B as "hominins" GD is asserting that they were not modern humans. I'm not certain whether DE originated in sub-Saharan Africa, north Africa or Asia, but it is certain that if DE originated in in northern Africa or Asia, there were also modern humans in sub-Saharan Africa at that time, and if DE migrated back to southern Africa, it mixed with modern humans, not "hominins".

terryt said...

"Both YDNA and mtDNA are subject to natural selection, including for the human species that apparently experienced several important gender-related and metabolism-related changes within a short period of time".

I very much doubt that the 'selection' of haplogroup lines has very much at all to do with metabolism-related factors. To me the distribution and expansion of human haploid lines is the result of technological and cultural expansions.

"Changes may have radiated from somewhere else".

I'm sure they did. For example we now know that the immediate ancestors of Y-DNA P and Q originated and expanded from SE Asia. There must have been some major cultural or technological change that provided the impetus for such a major expansion. But, as GailT said with regard to another haplogroup, 'it mixed with modern humans, not "hominins"'.

Vincent said...

@Earl Snerd

You make a good point. The evidence of archaic admixture in sub-Saharan Africans suggests that the ancestors of modern farming and pastoralist populations (generally high in haplogroup E) did not have contacts with the ancestors of modern hunter-gatherers before 35 kya (indicated as the likely date of archaic introgression). The more one goes farther from Central and South Africa, the more tha archaic admixture decreases, with lowest observed admixture in West and East Africa (I exclude North Africa because there is possibly no archaic admixture there other than from Neanderthals). This means that haplogroup E was not African in origin, because if this was the case we would observe similar levels of admixture in all sub-Saharan populations. This does not exclude the possibility of E-carriers migrating into South Africa prior to 35 kya without interbreeding with indigenous groups (Hofmeyr skull, exhibiting strong metric and non-metric similarity to UP Europeans).

Therefore I suggest that most of the African non-Paleoafrican components cited by you originated after 35 kya (i.e. after interbreeding between E-carrying Eurasians and A- and B-carrying Paleoafricans), two of the non-African components (Maghrebi, Ethio-Somali) being the result of back-migrating E sub-clades from the Levant.

@terryt

You are welcome.

Unknown said...

It is quite clear that there is a great migration out of Europe, much earlier than any of the historically important had happened. The Pyramids and the kurgans are just the same **** in a larger package than the dolmen. It started in Europe and spread to Somalia, the Caucasus, israel, india and Korean. If this is not a sign of migration, which then would be enough proof. Of course dolmens are religious symbol, but what? It does not belong here.

Unknown said...

"The best scientific hypothesis is that a modern Eurasian population carrying Y-DNA DE lineages entered Sub-Saharan Africa and admixed with local hominin populations carrying hgs A00, A and B."

German, what is your intention in using the word hominiN (which includes the great apes) in contrast to "modern Eurasians" in the above sentence. Are you intending to imply that the Y haplogroups A and B not modern humans but are pre-HominiDs haplogroups? Because that is how it reads to me.

Genetically it is very clear that A and B are well within the HominiD group, intimately related to the other modern human haplogroups and only distantly related to cousin hominids like Neanderthals.

The oldest modern skulls that I know of are 160k years old from Africa (Herto, Ethiopia). This is well withing the timeframe for modern humans (~200kya) I think Y haplogroup ages are underestimated, but even the BT split is currently dated to only about 76kya. These definitively MODERN, AFRICAN humans must have been A or something similar.

German Dziebel said...

@Annie Mouse

"German, what is your intention in using the word hominiN (which includes the great apes) in contrast to "modern Eurasians" in the above sentence. Are you intending to imply that the Y haplogroups A and B not modern humans but are pre-HominiDs haplogroups? Because that is how it reads to me."

It reads to me in the same way: Y-DNA hgs A and B were absorbed into the modern human genetic pool from pre-existing, "archaic" African hominins.

"Genetically it is very clear that A and B are well within the HominiD group, intimately related to the other modern human haplogroups and only distantly related to cousin hominids like Neanderthals. "

How come? They are most divergent among haplogroups found among modern humans and not found outside of Africa along any of the putative out-of-Africa migration routes. People have no problem proposing that A00 is an introgressed lineage (or treat Polynesian hg C2 as introgression from an extinct Papuan substrate), so why A and B should be treated differently?

"The oldest modern skulls that I know of are 160k years old from Africa (Herto, Ethiopia). This is well withing the timeframe for modern humans (~200kya) I think Y haplogroup ages are underestimated, but even the BT split is currently dated to only about 76kya. These definitively MODERN, AFRICAN humans must have been A or something similar."

We don't have any ancient DNA evidence for this, skull evidence is unreliable and since it's unlikely that we would ever obtain any secure evidence from Africa because of DNA perishability there, I would explore other hypotheses. It's a dead-end.

terryt said...

"Are you intending to imply that the Y haplogroups A and B not modern humans but are pre-HominiDs haplogroups? Because that is how it reads to me".

As I understand his belief that is exactly what he means. But oddly although he claims Amerindians mixed with 'West Eurasians' and 'Papuans' he doesn't see these as 'pre-HominiDs'. German's beliefs are a very strange mix.

terryt said...

"The oldest modern skulls that I know of are 160k years old from Africa"

German does actually have a point. The data as we have it at present does suggest that A00 split from the remaining surviving Y-DNA lines before 'modern humans' first appeared. The question then arises as to at what point did those remaining lines become modern human? And how can we tell a modern human from its immediate ancestor?

It turns out that we can draw the dividing line between archaic and modern humans almost anywhere we wish to. There is certainly no abrupt change:

http://www.unz.com/gnxp/whole-genomes-as-a-window-into-the-past/

Adding to the problem is the apparent anomaly that the surviving Y- and mt-DNA lines each originate in separate regions within Africa, and at separate times: Y-DNA earlier and further west that mt-DNA. We don't even know if the two lines left Africa together.

However we do know that genes survive from at least two branches of modern humans' archaic Eurasian relations. Does this mean that individuals carrying Y-DNA from an 'archaic' human are less 'modern' than are those individuals carrying a-DNA from archaic Neanderthals or Denisovans?

terryt said...

"They are most divergent among haplogroups found among modern humans and not found outside of Africa along any of the putative out-of-Africa migration routes".

German, stop talking rubbish. You always insist that any population that moved out of Africa would automatically contain every single haplogroup that existed in Africa at the time. To me that shows you know very little about genetics or evolutionary biology. You even seem to have a very tenuous grasp of human behaviour. Did you buy your PhDs in anthropology? I agree Y-DNAs A and B are the most divergent from the 'main' stem of surviving human Y-DNAs. Of course it is impossible to be absolutely sure because no ancient Y-DNA has yet been discovered but the mt-DNA evidence supports the idea that
Y-DNA A and B are far closer to those Y-DNAs than they are to any Neanderthal of Denisovan Y-DNA lines.

"Y-DNA hgs A and B were absorbed into the modern human genetic pool from pre-existing, 'archaic' African hominins".

Surely the fact that A and B branch off early from the other surviving Y-DNAs is adequate evidence for a deep African origin for all surviving Y-DNAs.

"(or treat Polynesian hg C2 as introgression from an extinct Papuan substrate)"

More rubbish. There is nothing 'extinct' about Y-DNA C2 (or C1c as it is now classified). The basal form of the haplogroup is still found in Southern Wallacea, where it almost certainly originated.

GailT said...

The data as we have it at present does suggest that A00 split from the remaining surviving Y-DNA lines before 'modern humans' first appeared.

The age estimates for A and B are well within the estimates for the origins of AMH, so there is no basis for GD or anyone else to claim that they represent archaic humans.

A00 is dated much more recently than the split between Neandertals and AMH, so even if A00 did represent an archaic human line (which is questionable given the uncertainty in the dates for A00 and for the origin of AMH), it was still more modern than Neandertals.

German Dziebel said...

@TerryT

"You always insist that any population that moved out of Africa would automatically contain every single haplogroup that existed in Africa at the time."

A bottleneck is a special explanation that requires special proof. I'm not aware of any special proof for an out-of-Africa bottleneck. It's just assumed. We easily find A00, A, B and E in African Americans and they represent a well-documented migration out of Africa since 1492. No magical bottleneck made a tiny subset of the African population which was made slaves leave all of their African genes behind. They are all well and alive in the New World now. Nothing of that sort is seen along the putative route(s) of an out-of-Africa migration 100,000-40,000 years ago. Hence, A00, A and B were likely absorbed from "archaic" hominins.

"Did you buy your PhDs in anthropology?"

No. I've earned two of those. What's the ROI on your high school diploma, Terry?

@GailT

"The age estimates for A and B are well within the estimates for the origins of AMH,"

Where did we get those estimates? From a bunch of poorly preserved and half-archaic skull fragments not associated with any modern human behavior? This is not a very secure material to create a baseline to interpret everything else.

terryt said...

"The age estimates for A and B are well within the estimates for the origins of AMH, so there is no basis for GD or anyone else to claim that they represent archaic humans".

Exactly, but that still leaves open the status of A00. The fact that the non-African Y-DNA branch, CT, forms a closely related branch to African B is surely evidence that those non-African haplogroups originated in Africa. Therefore A and B did not 'introgress' into the 'modern human genetic pool'. They lie at the base of it.


"A00 is dated much more recently than the split between Neandertals and AMH"

Yes. Even if A00 represents an 'archaic' line it is far closer to the modern human lines than it is to any other archaic population.

astenb said...

@ Vincent. Looking at these two statements:

"The evidence of archaic admixture in sub-Saharan Africans suggests that the ancestors of modern farming and pastoralist populations (generally high in haplogroup E) did not have contacts with the ancestors of modern hunter-gatherers before 35 kya"

and

"Therefore I suggest that most of the African non-Paleoafrican components cited by you originated after 35 kya (i.e. after interbreeding between E-carrying Eurasians and A- and B-carrying Paleoafricans)"

Are void of logic. Or perhaps are based on your ignorance of African genetics. You are running into the same problem of knowledge as German.

First of all you make the mistake of thinking Farming and pastoral populations are defined by E. AFRICANS as a whole are high in E. But the African ethnic groups that have the longest history of pastoralism are actually High in Haplogroups A and B. In fact some of them are A and B exclusive in Southern Sudan. These would be speakers of the Nilo-Saharan phylum. Keep in mind the pastoralism precedes agriculture in Africa by arguably 4 or 5 THOUSAND years. The Transmittal of pastoral practices and technology as well as that language family has been associated with Haplogroups A and B.

Secondly your other idea about pastoral and Farmer ancestries not having "contact" is also dubious because in the case of Haplogroups A and B. They actually have SHARED ancestors reaching back far before 35kya.

-A-M13 - Is the lineage that defines the root haplogroup of East and Central Africans, Nilo-Saharan and Afroasiaitc alike. On the other hand A-M51 defines the main root lineages in Southern African Hunter-gatherers. Both are united by A-M32 in DEEP antiquity...way before 35kya. Their ancestry is in fact shared on this Y-chromosome. Southern Africans also carry A-M6.

This observation was highlighted in 2001 in an article hence the Title :
Ethiopians and Khoisan Share the Deepest Clades of the Human Y-Chromosome Phylogeny. http://www.ncbi.nlm.nih.gov/pmc/articles/PMC384897/

-Moving on the Haplogroup B it shows the exact same pattern. B-M109 and B-G1 define the main B lineages in Nilo-Saharans and Bantu of Eastern and Central Africa. Meanwhile B-M112 is the main Haplogroup lineage associated with Twa (Pygmies) and Hunter-gatherers in East and Central Africa. There is a very deep divergence between the two but they SHARE mutation B-M812 prior to 35kya.

Lastly the entire idea is bankrupt because you are ignoring the maternal data indicating long term continual migration.

Vincent said...

Mr akb, thank you for the compliments and good words spoken to me, I am honoured. Now let's correct something.

"First of all you make the mistake of thinking Farming and pastoral populations are defined by E."

How is this a mistake? Bantu people occupy the majority of Central and South Africa. Bantu people have expanded considerably due to farming techniques. 1+1 = Bantu people were and are farmers, together with West Africans, from which they split. Is this a mistake? I don't think so.
I said "generally high in haplogroup E", not "exclusively", learn how to read. Haplogroup E-P2 is the most correlated both to farming populations and pastoralist populations (see Gebremeskel et al. 2014). Pastoralist populations are higher in haplogroup E than in other haplogroups, except in Sudan.

"AFRICANS as a whole are high in E"

Yes, especially non-hunter-gatherers (a.k.a. Paleoafricans).

"Secondly your other idea about pastoral and Farmer ancestries not having "contact..."

Learn how to read, Part Two. Where did I state "pastoral and Farmer ancestries not having contact"? I stated that the ANCESTORS of pastoralist AND farmers (i.e. haplogroup E and its sub-clades) not having contact WITH HUNTER-GATHERERS prior to 35 kya. You didn't get my point, that is: Neo-Africans (haplogroup E) back-migrated from Eurasia and did not have contacts with Paleoafricans (haplogroups A and B) prior to 35 kya, because this date marks the interbreeding between Paleoafricans and an archaic human form.

"Lastly the entire idea is bankrupt because you are ignoring the maternal data indicating long term continual migration."

And, can you tell me how does this correlate with my "ideas"? I find it difficult to get the meaning out of this sentence of yours, but I answer the way I understood it. Haplogroup E caused the extermination of the male Paleoafrican population, then its carriers took Paleoafrican women. It's like R-V88 in Northern Cameroon: paternal lineage of Eurasian origin, maternal lineages of exclusively Paleoafrican origin.

That said, do you have other "evidence" to attack the rest of what I wrote? If yes, first go back to your parents and ask them to educate you better. Education is important when you discuss with other people. The problem of knowledge is more yours than mine.

terryt said...

"A bottleneck is a special explanation that requires special proof. I'm not aware of any special proof for an out-of-Africa bottleneck. It's just assumed".

German, do you really believe that haplogroups A00, A0, A0a, A0b, A1, A1a, A1b, A1b1 and A1b2(in fact BT) were evenly distributed across the whole African continent when any population emerged into Eurasia? If find such a scenario impossible to accept, but obviously you believe every modern population contains exactly the same proportion of haplogroups as every other one.

"We easily find A00, A, B and E in African Americans and they represent a well-documented migration out of Africa since 1492"

Or are you now claiming that the 'migration' to America left from exactly the same region as the postulated out of Africa one? You childish clinging to unreality is perhaps to be admired, but it certainly does not involve 'science'.

"From a bunch of poorly preserved and half-archaic skull fragments not associated with any modern human behavior?"

You are now making the mistake ( and I readily admit you are not the only one) of confusing technological developments with genetic ones. 'Modern human behaviour' is an almost totally separate, and later, development from 'Modern human phenotype'.

German Dziebel said...

@akb

"Are void of logic. Or perhaps are based on your ignorance of African genetics. You are running into the same problem of knowledge as German.

First of all you make the mistake of thinking Farming and pastoral populations are defined by E. AFRICANS as a whole are high in E. But the African ethnic groups that have the longest history of pastoralism are actually High in Haplogroups A and B. In fact some of them are A and B exclusive in Southern Sudan."

You are just another nameless worshipper of Africa, the womb of peoples. The distribution of hgs A and B in Africa is irrelevant to the point I (and maybe Vincent) am making. Compare: Maoris are 80+% hg C2, which branched off from a node far more upstream than hg O1, which is frequently found in Taiwan. By your "logic," Austronesians came from New Zealand and colonized Taiwan. The reality is the opposite. Why? Because hg C2 was picked up from an extinct "Papuan" substrate somewhere along the way from taiwan to New Zealand. Just like hgs A and B were picked up from extinct hominin species in Africa by the incoming carriers of hg DE, which is a subset of the extra-African and modern human clade CF. Since no hgs A00, A or B are found along the putative out-of-Africa route of migration of modern humans (but they are abundantly found among post-1492 African Americans, so we know what genetic signature a true out-of-Africa migration would leave behind), modern Africans must have picked them up from a substrate. BTW, Ust-Ishim at 45K is not hg A or B, so no African-specific Y-DNA lineages have so far been detected in ancient remains outside of Africa.

terryt said...

"hg C2 was picked up from an extinct 'Papuan' substrate somewhere along the way from taiwan to New Zealand".

Why do you pretend to be stupid? Or are you not pretending? I have already pointed out that your Papuan substrate is far from 'extinct'.

" Since no hgs A00, A or B are found along the putative out-of-Africa route of migration of modern humans (but they are abundantly found among post-1492 African Americans, so we know what genetic signature a true out-of-Africa migration would leave behind)"

Again either pretending to be stupid or actually being stupid. The 'putative out-of-Africa route of migration' would have been through the extremely narrow stretch of land between the Mediterranean and Red Seas. The 'post-1492' migration involved the whole west African coast from the Gambia to Angola, and points inland. It is hardly surprising that just a single branch of Africa Y-DNA BT emerged through the first gap while a wider sample made the second exit. Again your comparison is invalid.

"Ust-Ishim at 45K is not hg A or B"

If you know what it is please tell.

German Dziebel said...

@terryT

"German, do you really believe that haplogroups A00, A0, A0a, A0b, A1, A1a, A1b, A1b1 and A1b2(in fact BT) were evenly distributed across the whole African continent when any population emerged into Eurasia? If find such a scenario impossible to accept, but obviously you believe every modern population contains exactly the same proportion of haplogroups as every other one. "

Recent discoveries have made even the dogmatic geneticists question the serial bottleneck scenario for modern human population dispersals (http://biorxiv.org/content/early/2014/03/21/003517), so I'm not alone in questioning it. I take this kind of critique to the next level by pointing out that a bottleneck is a special case that needs special proof. Lineages that are postulated to be the oldest had more time to spread around and considering that A00 coalesced at some 300,000 YBP and CDEF presumably left Africa 50,000 YBP all those African-specific lineages should've been present all over Africa by then. The chances to pick up older, more widely spread lineages are higher than the chances to pick up the newly evolved ones. Considering that hgs D, C and F are not found in Africa, we don't have any evidence that they evolved there. Considering that African-specific A00, A0, A0a, A0b, A1, A1a, A1b, A1b1 and A1b2 lineages are not found outside of Africa along the putative routes of the founding migration, an African population apparently never left Africa. The necessary conclusion is that Africa was peopled from Eurasia and A00, A0, A0a, A0b, A1, A1a, A1b, A1b1 and A1b2 were absorbed from an extinct substrate.

"Or are you now claiming that the 'migration' to America left from exactly the same region as the postulated out of Africa one?"

East Africa is even more diverse than West Africa - all the more reasons to expect all those African-specific lineages to be dispersed around Eurasia and the Sahul.

"'Modern human behaviour' is an almost totally separate, and later, development from 'Modern human phenotype'."

Again, it's a special hypothesis that requires special proof which is missing. There are no attested human populations who are modern phenotypically but not behaviorally, so "AMH" is a myth. Archaeology is not a good science to identify human behavior, hence the only reliable sources are worldwide patterns of modern behavior trait distribution found in linguistics, kinship studies, ethnology, etc.

GailT said...

@GD

By your "logic," Austronesians came from New Zealand and colonized Taiwan. The reality is the opposite. Why? Because hg C2 was picked up from an extinct "Papuan" substrate somewhere along the way from taiwan to New Zealand.

No, that is your conclusion based on your faulty logic and your complete ignorance of the methods of phylogeographic analysis.


Since no hgs A00, A or B are found along the putative out-of-Africa route of migration of modern humans (but they are abundantly found among post-1492 African Americans, so we know what genetic signature a true out-of-Africa migration would leave behind)

It is estimated that 11 million Africans survived the post-1492 slave transport to the Americas. They were captured from across sub-Saharan Africa. The fact that you claim this would present the same genetic signature as the migration of small bands of hunter-gatherers from northeast Africa ca 70,000 years ago proves either that your are trolling, or that you are an idiot.

German Dziebel said...

@GailT

"No, that is your conclusion based on your faulty logic and your complete ignorance of the methods of phylogeographic analysis."

You are just bluffing. I cited a well known fact.

"It is estimated that 11 million Africans survived the post-1492 slave transport to the Americas. They were captured from across sub-Saharan Africa. The fact that you claim this would present the same genetic signature as the migration of small bands of hunter-gatherers from northeast Africa ca 70,000 years ago proves either that your are trolling, or that you are an idiot."

It doesn't matter how many African slaves were captured or how many survived. What matters is that after 1492 a regionally circumscribed subset of a much larger African population moved to the New World. The genetic signature of this migration creates a gold standard of what we should expect to see if we postulate a subset of a much larger African population to have left Africa 50,000-60,000 years ago. In the former case, we see the presence of African specific lineages, including most "ancient" ones such as Y-DNA A00 present in the daughter population. We don't see anything of that sort anywhere along the putative route(s) of an out of African migration 50,000 years ago. You are welcome to try and prove, using ancient DNA, your ideas about a bottleneck that left all the most ancient African lineages behind but spread all the most recent ones all over the globe but for now they are baseless.

BTW, most African Americans trace their roots to West Africa (http://www.pnas.org/content/early/2009/12/10/0909559107.short?rss=1). So you need to educate yourself about the history of the world in the last 500 years before you venture any further.

Gail, you are a troll AND an idiot. And I'm very glad that you represent out-of-Africa. We need more of folks like you - makes out-of-Africa an easy target for highly trained professionals like myself.

terryt said...

"Recent discoveries have made even the dogmatic geneticists question the serial bottleneck scenario for modern human population dispersals (http://biorxiv.org/content/early/2014/03/21/003517), so I'm not alone in questioning it".

You are alone in questioning the African origin of modern haplogroups. The link questions the 'serial bottleneck scenario', not the 'out of Africascenario'. All that link says is that the spread beyond Africa was not simple. Populations both within Africa and outside it moved backwards and forwards. Quote:

"long-­‐range migration and concomitant population replacement
or admixture has occurred often
enough in human history that the
present-­‐day inhabitants of many
places in the world often bear
little genetic relationship to
the more ancient peoples of the
same region".

You are, once more, being deliberately stupid.

"Lineages that are postulated to be the oldest had more time to spread around"

'More time' is totally independent of 'more opportunity'. I'm sure you realise that and so again you're being deliberately stupid. You are certainly demonstrating ' your complete ignorance of the methods of phylogeographic analysis'.

"all those African-specific lineages should've been present all over Africa by then. The chances to pick up older, more widely spread lineages are higher than the chances to pick up the newly evolved ones."

They quite possibly were spread widely in Africa but subsequent movements of 'newly evolved ones' have eliminated them in many of the more accessible regions. And don't forget that 'newly evolved' Y-DNAs are only noticed because they have spread. Many new Y-DNA mutations arise but fail to lerave descendants.

"Considering that African-specific A00, A0, A0a, A0b, A1, A1a, A1b, A1b1 and A1b2 lineages are not found outside of Africa along the putative routes of the founding migration, an African population apparently never left Africa".

Another demonstration of 'your complete ignorance of the methods of phylogeographic analysis'. The fact that 'CDEF' is a clade within 'A1b2' surely shows that haplogroup is African in origin. And the fact we do not find a trail of African Y-DNA along any route out of Africa shows just a single branch emerged from that continent. Or any others that did so failed to survive.

"East Africa is even more diverse than West Africa"

So?

"all the more reasons to expect all those African-specific lineages to be dispersed around Eurasia and the Sahul".

Much of that East African diversity is a product of Eurasian lineages moving back to Africa. Another source is later arrivals of West African Haplogroups.

"Again, it's a special hypothesis that requires special proof which is missing".

The 'special proof' is surely required from those who claim 'modern behaviour' is a product of genetics. There is a huge variety of modern behaviour, none of which coincides with genetics.

terryt said...

"What matters is that after 1492 a regionally circumscribed subset of a much larger African population moved to the New World".

I get the distinct impression your comments regarding Africans is inspired by your deeply held racist attitude. We know that 'movement' did not come from 'a regionally circumscribed subset'. It came from virtually the whole west coast of Africa and even from parts of the east coast during the later stages of the slave trade.

"The genetic signature of this migration creates a gold standard of what we should expect to see if we postulate a subset of a much larger African population to have left Africa 50,000-60,000 years ago".

You're being deliberately stupid again, German. Surely the evidence shows that a much smaller population 'left Africa 50,000-60,000 years ago' than the population that left since 1492. And the earlier exit comprised a much smaller subset than did the later exit.

"most African Americans trace their roots to West Africa (http://www.pnas.org/content/early/2009/12/10/0909559107.short?rss=1)".

Yes, 'most'. From the link.

"African ancestry is most similar to non-Bantu Niger-Kordofanian-speaking populations, consistent with historical documents of the African Diaspora and trans-Atlantic slave trade".

'Most' of the ancestry of modern African-Americans derives from those people because they were the first there and have therefore provided most of the population.

"We need more of folks like you - makes out-of-Africa an easy target for highly trained professionals like myself".

So can you provide a list of 'highly trained professionals' who agree with your out of America belief?

GailT said...

The genetic signature of this migration creates a gold standard of what we should expect to see if we postulate a subset of a much larger African population to have left Africa 50,000-60,000 years ago.

Seriously? You actually believe that the population that expanded out of Africa 60,000 years ago was much larger than 11 million people? It is possible that it was no more than a few tens of thousands of people, or less, and they migrated from northeast Africa. It's absolutely crazy to suggest that there was a mass exodus from the entire continent 60,000 years ago that represented the full diversity of Africa at that time.

German Dziebel said...

@GailT

"Seriously? You actually believe that the population that expanded out of Africa 60,000 years ago was much larger than 11 million people? It is possible that it was no more than a few tens of thousands of people, or less, and they migrated from northeast Africa. It's absolutely crazy to suggest that there was a mass exodus from the entire continent 60,000 years ago that represented the full diversity of Africa at that time."

I've always doubted your analytical abilities, but now I begin doubting your reading comprehension abilities, too. Just re-read my paragraph. The point is that in both cases (post 1492 and 50,000 years ago) a subset of a larger population left (or is proposed to have left) Africa. In 1492 the size of the migrant population was larger and the size of the source population was larger. But in the case of 1492 we have a clear evidence for the most divergent and African-specific lineages being carried out by the migrants. In the case of 50,000 years ago we don't. Why? Likely because there was no migration out of Africa in the latter case. On the other hand, we do have evidence for Eurasian-specific human lineages and Eurasian hominin lineages in Sub-Saharan Africa. A logical conclusion would be that those highly divergent and African-specific lineages were not part of the original modern human gene pool but were absorbed by the carriers of Eurasian lineages upon their entry into Africa.

terryt said...

"In 1492 the size of the migrant population was larger and the size of the source population was larger. But in the case of 1492 we have a clear evidence for the most divergent and African-specific lineages being carried out by the migrants. In the case of 50,000 years ago we don't. Why?"

The evidence overwhelmingly indicates that just one Y-DNA African line and perhaps two African mt-DNA lines emerged on the earlier occasion.

"Likely because there was no migration out of Africa in the latter case".

Wrong. There obviously was a migration out of Africa because all surviving haplogroups derive from African ones.

"A logical conclusion would be that those highly divergent and African-specific lineages were not part of the original modern human gene pool but were absorbed by the carriers of Eurasian lineages upon their entry into Africa".

That doesn't make any sense at all. If the 'original modern human gene pool' was unable to breed with those 'highly divergent and African-specific lineages' before they left Africa the 'the carriers of Eurasian lineages upon their entry into Africa' would certainly have been unable to breed with them.

Anonymous said...

terryt posted:

"There obviously was a migration out of Africa because all surviving haplogroups derive from African ones."

Circularity is very circular.

German Dziebel said...

@Mangrove

"Circularity is very circular."

Bingo!