Thursday, June 5, 2014

Most Northern Hemisphere Men Have Patrilineal Ancestry In SE Asia

[W]e genotype 13 new highly informative single-nucleotide polymorphisms in a worldwide sample of 4413 males that carry the derived allele at M526, and reconstruct an NRY haplogroup tree with significantly higher resolution for the major clade within haplogroup K, K-M526.

Although K-M526 was previously characterized by a single polytomy of eight major branches, the phylogenetic structure of haplogroup K-M526 is now resolved into four major subclades (K2a–d). The largest of these subclades, K2b, is divided into two clusters: K2b1 and K2b2. K2b1 combines the previously known haplogroups M, S, K-P60 and K-P79, whereas K2b2 comprises haplogroups P and its subhaplogroups Q and R.

Interestingly, the monophyletic group formed by haplogroups R and Q, which make up the majority of paternal lineages in Europe, Central Asia and the Americas, represents the only subclade with K2b that is not geographically restricted to Southeast Asia and Oceania. Estimates of the interval times for the branching events between M9 and P295 point to an initial rapid diversification process of K-M526 that likely occurred in Southeast Asia, with subsequent westward expansions of the ancestors of haplogroups R and Q.
From Tatiana M Karafet, Fernando L Mendez, Herawati Sudoyo, J Stephen Lansing and Michael F Hammer, "Improved phylogenetic resolution and rapid diversification of Y-chromosome haplogroup K-M526 in Southeast Asia", European Journal of Human Geneticas (June 4, 2014).

Y-DNA R and Q have origins in SE Asia Probably Between 45 kya and 70 kya.

The most common Y-DNA haplogroup type in Europe is R (specifically, R1a and R1b), although this distribution probably didn't arise for most of Europe until the second wave of farmer/herder migrations to Europe, a thousand to two thousand years or so after the first farmers brought the Neolithic Revolution to Europe. 

R1a is associated with Indo-European settlement starting as far back as the Corded Ware culture or earlier in Central to Eastern Europe and Central Europe with ancient DNA evidence of the population's genotype extended all of the way to the Tarim Basin in far Western China at its greatest extent for essentially all of the Bronze Age and into the 7th century CE or so, when Uygur populations emerge in the region.

R1b is the predominant Y-DNA haplogroup of Western Europe and a very basal branch of it (R1b-V88) is common among the mostly pastoralist speakers of the Chadic languages in the African Sahel, a branch that probably arrived around 5,200 years ago based on archaeological evidence.  A migration probably brought this Y-DNA haplogroup into Western Europe at high frequencies sometime after that, although the precise archaeological culture associated with the predominance of R1b in Western Europe isn't entirely clear.  I've advanced that hypothesis that the Bell Beaker culture (the earliest to have R1b ancient DNA discovered so far) was responsible (for lack of any better candidates), but other evidence points to the prior Megalithic culture members who were first farmers in much of Western Europe, but with very fragile evidence, mostly from archaeologically inferred population booms and busts and no ancient DNA at all, to support that hypothesis.

Y-DNA haplogroup R2 has a largely South Asian distribution with concentrations highest in the Indus River Valley. Ancient DNA reveals an Y-DNA R* individual, Ma'lta boy ca. 24,000 years ago, with significant autosomal genetic affinity to modern Native Americans near the Altai region of Southern Siberia.

Y-DNA haplogroup Q is the most common Y-DNA haplogroup of Native Americans and is also found in Siberia.

Thus, the Y-DNA P clade within newly redesignated Y-DNA K2b is the leading Y-DNA clade of the Northern part of the Northern hemisphere.  The latest study cited above indicates that the Y-DNA P clade that includes R and Q probably arose in Southeast Asia, coincident with the expansion of Y-DNA K2, in general, and then migrated West together before expanding into their current range, from an intermediate source somewhere in the region from India to Iran to Central Asia.

More On The Expansion of K2 into SE Asia

Both Y-DNA haplogroup C (whose Asian specific break from other Eurasian Y-DNA haplogroups is more basal than K) and Y-DNA haplogroup P (which is part of K2) haplogroups are found in Australian aboriginal men and indigeneous Papuan peoples. So, the rapid expansion of K2 giving rise to sub-haplogroups had to have taken place before about 45,000 years ago, the earliest securely dated archaeological evidence of modern human presences in those places.

It isn't clear if C and K arrived in Southeast Asia in a single migration of a genetically mixed population, or if they were two separate waves of migration that only started to produce a Y-DNA haplogroup mixed population in SE Asia prior to migrations to Australia and Papua New Guinea.

Suggestive evidence, but no smoking evidence, suggests that modern humans most likely reached Southeast Asia from South Asia around the time of the Toba volcanic eruption ca. 74,000 years ago.  The 74,000 years ago date would also fit an mtDNA mutation rate estimated expansion of mtDNA haplogroups N and R out of India.

This time span (from 45,000 years ago to 74,000 years ago) is long enough to accommodate either a one wave or two wave scenario.  Time ranges from 100,000 years ago to about 55,000 years ago could be squared with hypothesis proving interpretations of the archaeological evidence.

My weak personal bias is an expectation that these waves of migration were separate, with C expanding first (probably around 74,000 years ago), based mostly upon the phylogeny of the Y-DNA C haplogroup expanded in Asia, which seems to show a rapid "race across the coastal route" path from India that differs quite a bit from the pattern seen in haplogroup K.  Haplogroup K2 would then migration to SE Asia in a separate wave perhaps 5,000 to 10,000 years later.

Also, sometime between 45,000 and 74,000 years ago, somewhere between India and the island of Flores, these early modern human migrants to Asia admixed with archaic hominins whose Denisovan-like genetic traces are present in modern Australian Aborigines, Melanesians and Phillipino Negrito populations at significant levels.  The peak admixture percentage in the source population for these groups is estimated to have been around 8%.  My weak personal bias that this admixture took place mostly on the island of Flores with H. Florensis who would have Denisovan DNA in that scenario, but I'm less confident of any particular scenario for Denisovan admixture than I once was, mostly because of (1) possible evidence of a low but non-zero level of mainland Asian Denisovan admixture that is highly diluted but was not eliminated that is unlikely to be due to Melanesian back migration, and (2) physical anthropology and archaeology suggestions that H. Florensis may have been Homo Erectus that evolved to a smaller size due to island dwarfism.  But, I have not yet seen really solid evidence of either point.

Homo Erectus was the first hominin species to leave Africa and arrived in Asia including South East Asia around 1,800,000 years ago.  It isn't clear if Homo Erectus was still in SE Asia at the time and went extinct as part of the same Toba/migration event, or if this hominin species went extinct in the region at an earlier time, because the archaeological record of them more recently than 100,000 years ago is pretty much non-existent.  The genetic evidence related to Denisovan admixture, strongly supports the concluding that the Archaic Denisovan admixture seen in modern humans is from a hominin species that probably was not a direct decedent of Asian Homo Erectus.

The species with Denisovan genetics might have replaced Homo Erectus before modern humans reached Southeast Asia from South Asia, or might have co-existed with Homo Erectus in Asian until modern humans brought about the extinction of Homo Erectus.  In either case, the extinction of Homo Erectus could have happened either directly, via genocide, or indirectly, for example, through impacts on food supply and habitat.  An extinction of Homo Erectus entirely, without a strong role in that process from another hominin species, seems unlikely for an archaic hominin species that had already managed to persist for 1,700,000 years or so in Asia.

A Toba/modern human migration model of Homo Erectus extinction seems more parsimonious.

There is no meaningful archaeological evidence of an intermediate archaic homin species other than Homo Flores (which seems like a poor candidate to wipe out a continent wide Homo Erectus population) apart from Denisovan admixture in the populations mentioned above, but we really don't have much solid evidence one way or the other to discriminate between different models of Homo Erectus extinction.

Footnote on Y-DNA Haplogroup D

Y-DNA haplogroup D is the other main Y-DNA haplogroup, in addition to C and K2 that is found in East Eurasia.  It is very common in Japan and is also found in a swath of land from the Anadaman Islands, across South Asia to Tibet and into the steppe to the North of Tibet.  The Japan v. non-Japan split in the phylogeny is more basal than the splits within each category and is almost complete (i.e. only Japanese D is found in Japan, and only non-Japanese D is found in South Asia).

The details of a DE split are controversial and beyond the scope of this post, but the question with Y-DNA haplogroup D is whether its island-like distribution is a result of a much wider range that was split up by subsequent migration, or whether it migrated to Asia after Y-DNA C and K2, perhaps seeking out niches not already occupied by modern humans in Asia ca. 30-40 kya.  I tend to favor the later scenario, because there is so little evidence of Y-DNA D anywhere else in Asia, and because the later scenario is still consistent with the evidence on the earliest migrations of modern humans to Japan.  I also tend to favor a scenario in which the Japanese Y-DNA D carriers arrive via a Northern route from somewhere North of Tibet, rather than a Southern Coastal route.

On the other hand, mutation rate data tend to favor a spit between Y-DNA D and Y-DNA E around the same time as the split between Y-DNA K1 and Y-DNA K2, for which the archaelogically favored date would be around 70,000 years ago, or perhaps a few thousand years earlier.  It could be that Y-DNA D expanded on a Northern route from Central Asia to Korea and then trickled down into India and the Andaman Islands from Tibet, prior to the LGM ca. 20,000 years ago, while Y-DNA C and K2 expanded along a Southern route from India into South East Asia, but that the LGM ice age wiped out all but a few relict populations with Y-DNA D, thus breaking up its range.

Another factor favoring Y-DNA haplogroup D as a post-Y-DNA C and K2 wave is that none of the Y-DNA haplogroup D populations has any significant non-Neanderthal archaic hominin admixture.

The New Y-DNA Haplogroup K1.

The Y-DNA haplogroups known as L (mostly South Asian and especially the Indus River Valley) and T (geographically broad early Neolithic and/or Epipaleolithic expansion from around Mesopotamia into Europe and parts of Africa and since diluted in many places), have been included in a newly designated Y-DNA K1 haplogroup.

It would appear that K1, which stayed in West Eurasia (probably in Iran or South Asia) before it split from K2, which continued on to SE Asia around 70 kya, if not earlier.

K1 could also have backmigrated from Southeast Asia around the same time as Y-DNA haplogroups Q and R, but that seems much less likely as there are few traces of Y-DNA haplogroup L or T in Southeast Asia that can't be much more easily explained by later migrations.





13 comments:

barakobama said...

"Y-DNA haplogroup R2 has a largely South Asian distribution with concentrations highest in the Indus River Valley. Ancient DNA reveals an Y-DNA R* individual, Ma'lta boy ca. 24,000 years ago, with significant autosomal genetic affinity to modern Native Americans near the Altai region of Southern Siberia."

The dude was west Eurasian, which makes native Americans part west Eurasian, I don't know why some people still don't understand this.

terryt said...

"I don't know why some people still don't understand this".

I think most do understand it. It's just German who has such a problem coming to grips with the idea.

"Both Y-DNA haplogroup C (whose Asian specific break from other Eurasian Y-DNA haplogroups is more basal than K) and Y-DNA haplogroup P (which is part of K2) haplogroups are found in Australian aboriginal men and indigeneous Papuan peoples".

P is not found in Australian Aborigines. Just K-P60 in Australia, shared with New Guinea.

"So, the rapid expansion of K2 giving rise to sub-haplogroups had to have taken place before about 45,000 years ago, the earliest securely dated archaeological evidence of modern human presences in those places".

That comment is only correct if it is assumed the two haplogroups arrived in Australia simultaneously. I don't think that assumption is justified. I think we can safely assume Australia would have been settled from Timor before New Guinea was. K is much more heavily represented, and diverse, in New Guinea/Melanesia than in Australia whereas C is virtually absent in New Guinea/Melanesia, and the branch that is there (C2a) is very much associated in some way with the Austronesian expansion.

"My weak personal bias is an expectation that these waves of migration were separate, with C expanding first"

Yes.

"Suggestive evidence, but no smoking evidence, suggests that modern humans most likely reached Southeast Asia from South Asia around the time of the Toba volcanic eruption ca. 74,000 years ago. The 74,000 years ago date would also fit an mtDNA mutation rate estimated expansion of mtDNA haplogroups N and R out of India".

To me the phylogeny of mt-DNA N and Y-DNA C is inconsistent with any original expansion through India for either haplogroup.

"It could be that Y-DNA D expanded on a Northern route from Central Asia to Korea and then trickled down into India and the Andaman Islands from Tibet"

That's what I think happened. And if D could do it, so could C. C's southern element is entirely a single branch, C1, whose relation, C2, is most definitely East Asian. Therefore it is most likely that C1 originated somewhere in that region too.

"It [Y-DNA D] is very common in Japan and is also found in a swath of land from the Anadaman Islands, across South Asia to Tibet and into the steppe to the North of Tibet".

Not exactly 'across South Asia'. Almost entirely confined to the northeast. And certainly nothing 'coastal' about that distribution either. Although it obviously did reach the coast at either end of its almost completely inland distribution.

"I also tend to favor a scenario in which the Japanese Y-DNA D carriers arrive via a Northern route from somewhere North of Tibet, rather than a Southern Coastal route".

As did Tibetan D. And Andaman D almost certainly arrived after moving south from near Tibet.

"but that the LGM ice age wiped out all but a few relict populations with Y-DNA D, thus breaking up its range".

Exactly. It split D from E and C from F, with the second of each pair western and the first eastern.

"K1 could also have backmigrated from Southeast Asia around the same time as Y-DNA haplogroups Q and R, but that seems much less likely as there are few traces of Y-DNA haplogroup L or T in Southeast Asia that can't be much more easily explained by later migrations".

I agree completely with that. LT is much more likely to have 'peeled off' on the route east. The same for IJ yet further west.

terryt said...

Correction:

"C is virtually absent in New Guinea/Melanesia, and the branch that is there (C2a)"

Sorry. That is old nomenclature. The main New Guinea branch is now called 'C1c1-M208' although a minor level of C1c-M38 is also present, and presumably accompanied its descendant haplogroup o its voyage from southern Wallacea to western New Guinea, and beyond.

J. Lyon Layden said...

I am pretty much in agreementbwith you on this one except that I think soloensis might be a better choice than flores, and also may have both shielded flores from hss and later absorbed them into their population as hss pushed them into flores.
Yes the 27k dates for soloensis have been questioned but it makes sense that they were thgere considering hss fossils show up in eastern indonesia but not western indonesia all through the late paleolithic.
The soloensis goit pushed down into the java area of sundaland while hss traded from south china through sunda down to australia.
What i havent figured out yet is who the ghost population between papuan m and south asian LT was. Was it LT itself? Was it unmutated P, after assimilating the last of soloensis? Was it one of the clades ok K? Or was it soloensis themselves, and their genes have just gone extinct over the past 10,000 years. Someone had to spread aspm d...that someone was between the kalash and druze on ine side and the papuan agriculturists on the other.

terryt said...

"I think soloensis might be a better choice than flores"

Agreed. I think Flores 'Hobbits' survived after modern human arrival (if they did in fact survive, seems some doubt now) because the two groups split the ecology. The Hss that reached the island were probably coastal adapted.

"What i havent figured out yet is who the ghost population between papuan m and south asian LT was. Was it LT itself?"

Have you checked out the latest revision of Y-DNA? Seems K split into subhaplogroups as it spread out along the Sunda Islands with P first developing east of Wallace's Line before its huge expansion back west.

J. Lyon Layden said...

Yes i think it could be one of those Ks. P to me is tricky though....by east do you mean phillipines? Phillipines is east of the wallace line in the aeta but phillipines is also close to mainland asia. Traces of P are also found up the east coast of mainland asia. It would be nice if P would stay a north half of sundaland thing and spread out from there. If its really in new guinea or australia it makes the exodus more difficult with the whole aspm thing...or it might mean that a branh of P evolved aspm d after the ancestors of Q made their exodus. R has less aspm d than M or L or T, so they cant be the orinal source.

terryt said...

"by east do you mean phillipines?"

No. Probably in the triangle of islands formed by Timor, Sumba and Flores/Alor. In other wordfs not as far east as 'If its really in new guinea or australia'.

http://dienekes.blogspot.co.nz/2014/06/refined-structure-in-haplogroup-k-m526.html

My comment there related to Maju's post on the subject, 'Every single haplogroup in K2b has its major presence in Wallacea (counting the Philippines as Wallacean as they are certainly not Sunda), not in Sunda as he shows'. The other K2 haplogroups string out along the Sunda Islands.

"it might mean that a branh of P evolved aspm d after the ancestors of Q made their exodus. R has less aspm d than M or L or T, so they cant be the orinal source".

Neither Q nor R formed until long after P had left Wallacea.

J. Lyon Layden said...

Hmmm...very troublesome. If P is north, south,and east of soloensis and the homeland of k2d, k2e, and k* then what connects L or LT to M? Has to be something....even if we usethe kind of cop-outs that scientists use (genetic drift and founder effect) it still seems screwy. LT and M had to be adjacent to each other around the time of the LGM or shortly after...or M and L absorbed some ghost population coming out of sundaland as it sliwly flooded...a huge population that left no trace considering %60 + aspmd in kalah, druze, and papyans but nowhre near that high nywhere else.

J. Lyon Layden said...

Err..%60 + in kalash, druze, papuans

J. Lyon Layden said...

Right I agee didnt come from Sunda. I think it spent the last lgm in a central asian refugium, pretty much drove any other haplogroups in there with it to extinction. There is a 100,000 ybp allele that has made a sweep similar to Rs, and since R split off only 30k ago im thinking it introgressed via a ghost pop of pre- ED/CF humans or possibly another archaic...but itwas a seperate introgression than Micro-D or ASPM-D. Of course aspm-d may not be an introgression but a beneficial mutation to help us deal with Micro-D.

J. Lyon Layden said...

That one is DXS225...has the same distribution as ancient R would have had before the spread of NO and later Genghis Khan. Theres a strange absence of both in a little pocket of north italy, holdout of the Aurignatians or Gravettians. In australia, ng, and america it follows they follow the same post colonial distribution.

J. Lyon Layden said...

Dxs255 is passed from women to daughters and sons but from men only to daughters which also explains its distribution compared to R.
The more I look at this the more I think K1 may have remained isolated from k2 sonewhere near Java and built Gunang Padang before travelling up the river ne when sunda flooded. Then se from the holy mountain and down into pakistan to spread farther west. But then R came down south from central asia with horsed archery and steel.

J. Lyon Layden said...

K1 and subclades L and T are found in southeast asia though they rarely cgeck for L there, according to wikipedia.