December 11, 2009

Mapping Human Genetic Diversity in Asia

From the press release:
Several genome-wide studies of human genetic diversity have been conducted on European populations. Now, for the first time, these studies have been extended to 73 Southeast Asian (SEA) and East Asian (EA) populations.

In a paper titled, "Mapping Human Genetic Diversity in Asia," published online Science on 10 Dec. 2009, over 90 scientists from the Human Genome Organisation's (HUGO's) Pan-Asian SNP Consortium report that their study conducted within and between the different populations in the Asia continent showed that genetic ancestry was highly correlated with ethnic and linguistic groups.

The scientists also reported a clear increase in genetic diversity from northern to southern latitudes. Their findings also suggest that there was one major inflow of human migration into Asia arising from Southeast Asia, rather than multiple inflows from both southern and northern routes as previously proposed. This indicates that Southeast Asia was the major geographic source of East Asian and North Asian populations.

(A figure illustrating the paper shows plausible routes of pre-historical migration of Asian human populations. According to the study, the PanAsia SNP Initiative, the most recent common ancestors of Asians arrived first in India and later, some of them migrated to Thailand, and South to the lands known today as Malaysia, Indonesia, and the Philippines. The first group of settlers must have gone very far south before they settled successfully. These included the Malay Negritos , Philippine Negritos , the East Indonesians, and early settlers of the Pacific Islands. Thereafter, one or several groups of people migrated North, mixed with previous settlers there and, finally, formed various populations we now refer to as Austronesian, Austro-Asiatic, Tai-Kadai, Hmong-Mien, and Altaic. The figure is titled, "Putative Pre-Historical Migration Routes of Asian Human Populations.")

The researchers noted that the geographical and linguistic basis of genetic subgroups in Asia clarifies the need for genetic stratification when conducting genetic and pharmacogenomic studies in this continent, and that human genetic mapping of Asia has important implications for the study of genetics and disease and for research to understand migratory patterns in human history.

HUGO President Edison Liu, M.D., who is Executive Director of the Genome Institute of Singapore (GIS), said, "This study was a milestone not only in the science that emerged, but the consortium that was formed. Ten Asian countries came together in the spirit of solidarity to understand how we were related as a people, and we finished with a truly Asian scientific community. We overcame shortage of funds and diverse operational constraints through partnerships, good will, and cultural sensitivity.

"Our next goal is to expand this collaboration to all of Asia including Central Asia and the Polynesian Islands," said Dr. Liu, one of the corresponding authors of the paper. "We also aim to be more detailed in our genomic analysis and plan to include structural variations, as well as over a million single nucleotide polymorphisms in the next analysis."

While HUGO initiated and coordinated the research, Dr. Liu pointed out, "Affymetrix, led by Dr. Giulia C. Kennedy and based in the US, is our primary technology partner in this endeavour. We greatly appreciate their support."
As usual for a Science paper there are voluminous (and free) supplementary materials. The Neighbor-Joining tree shows the clear correlation between linguistic affiliation and genetic identity of individuals.


UPDATE: As to the main thesis of the paper, namely that East Asians are descended from Southeast Asians rather than Central Asians, I have to say that I am not convinced. This thesis is based on two observations: minimum sharing between East Asians and Central/South Asians and south-north reduction of genetic diversity in East Eurasians. However, the high genetic diversity in Southeast Asians can be explained if they are taken to be old hybrids of Mongoloid northerners with "Australoid"-like southerners as physical anthropology suggests, and the seeming absence of influence of present-day Central/South Asians is due to the fact that the latter are largely Caucasoids of western Eurasian origin, and, thus, do not represent any putative ancestral populations to modern Mongoloids.

Science
doi:10.1126/science.1177074

Mapping Human Genetic Diversity in Asia

The HUGO Pan-Asian SNP Consortium

Abstract

Asia harbors substantial cultural and linguistic diversity, but the geographic structure of genetic variation across the continent remains enigmatic. Here we report a large-scale survey of autosomal variation from a broad geographic sample of Asian human populations. Our results show that genetic ancestry is strongly correlated with linguistic affiliations as well as geography. Most populations show relatedness within ethnic/linguistic groups, despite prevalent gene flow among populations. More than 90% of East Asian (EA) haplotypes could be found in either Southeast Asian (SEA) or Central-South Asian (CSA) populations and show clinal structure with haplotype diversity decreasing from south to north. Furthermore, 50% of EA haplotypes were found in SEA only and 5% were found in CSA only, indicating that SEA was a major geographic source of EA populations.

Link

18 comments:

Anonymous said...

From the little I have read I have to say it appears to be a well structured and thought out study of ethnic and linguistic diversity in eastern part of Asia including parts of the Oceania abutting SE Asia.

The conclusion you can take with or without a grain of salt. Genetic diversity increases in large populations. Indonesia, SE Asia have large populations fed by rice grown in ideal conditions for that grain. Rice is the staple grain in most of Asia including the northern parts like Japan, North China that don't have ideal conditions for the growing of rice. Rice is also important in India. It would naturally follow that the equivalent of the West Eurasian Neolithic with wheat plus domestic animals from the Middle East occurred in the eastern parts of Asia with rice, water buffalos and their domestic animals. SE Asia would have provided a large part of the population that went on to found the present day populations of north and northeast Asia.

It is accepted that the first or primal movement of humans out of Africa went via the South Arabian route through India to southeast Asia. The movement of these hunter and gatherers stayed in some places like India, and moved on, some going to both North Asia and the SE Asia and islands of Oceania. That would account for most of the Australoid peoples. A later migration of humans who had become Mongoloid would have migrated from a source in SE Asia towards the north and south encountering and partially absorbing the Australoid peoples who preceded them.

If the aim of the study was to prove a SE Asian origin for most east Asians as distinct from a West Eurasian or South Asian one then it has succeeded in establishing that fact but not proving it outright. Present day SE Asian have a lot of South Asian admixture, not ancient as in many thousands of years but still important. It is reflected in the languages of SE Asia, the cultures, the religions and the appearance of SE Asians. Cambodians are the most admixed, followed by Malays and Thais. Naturally the mongoloid and to a lesser extent the Australoid would predominate, so that Cambodians, Thai, Indonesians, Malays and some Chinese minorities would cluster together.

The southern east Asian are more diverse for the reasons outlined: being an original source of humans immigrations to the north and south; admixing between mongoloid and Australoid human types; the fount of east Asian food crops, start of agriculture; and having a large population.

I always suspect the motives for genetic studies. There is an effort to totally diminish any contribution of West Eurasians to the populations of Asia. A bit of bigotry.

eurologist said...

Dienekes,

I agree with your: As to the main thesis of the paper, namely that East Asians are descended from Southeast Asians rather than Central Asians, I have to say that I am not convinced.

There is clearly a lack of populations today to prove the point that Mongoloids originate from central Asians and contributed independently from the other two populations to modern East Asians.

For India, the problem is not only that they are now dominated by Caucasoids, but also that (IMO) northern India/Pakistan was the original source population of Caucasoids - so their similarities go back 45,000 years with continued exchange. Note that climatically and culturally (modes of sustenance), northern India/Pakistan was somewhat separated from the south - even in the absence of mountain ranges, by a wide swath of barren land until after LGM.

With Mongoloids, it appears to me that even if they originated from (perhaps a Northeastern) Indian population, they clearly must have been much more isolated after they left.

The entire area between the Altai mountains and Beringia could of course account for such an isolation. My main dilemma is that I find it easier to get there from northern modern Afghanistan, Tajikistan, and Kyrgyzstan than via Northern Myanmar/ West China etc. However, the Mongoloid haplogroups seem to prefer the second route.

So, would it make sense to compare e.g. NE Indian, Bhutan, Northern Myanmar, and the Western Chinese populations with those of Mongolia, NE Siberia, Japan, Korea to look for a separate Mongoloid origin?

Maju said...

My main dilemma is that I find it easier to get there from northern modern Afghanistan, Tajikistan, and Kyrgyzstan than via Northern Myanmar/ West China etc. However, the Mongoloid haplogroups seem to prefer the second route.

The key issue is that humans, as tropical-adapted animals must have strongly prefered the tropical and subtropical areas of Asia to the far north. Also these warmer areas allow for much greater densities, so they can more easily become the source of new migrations.

Additionally West and East Eurasia are very much separated by deserts except at the narrow Altaian corridor and in South/SE Asia but only East of the Himalaya.

So SE Asia (including Southern China and NE India) must have been a major hub for the colonization of East Asia as a whole. Genetics support it, the limited archaeological data does not oppose it either and the only obstacle are in fact old and rather obsolete and somewhat mythological (speculative, idealist) anthropometry-based theories of many decades ago.

For me it's pretty clear: the main migratory flow was from South/SE Asia, with much emphasis for SE Asia, the same (but also rather distinct after some moment) as the one to Sunda/Wallacea/Sahul. This flow pre-dates the formation of the Mongoloid morphotype(s), which is a process of regional homogenization that belongs to East Asia only after divergence from "Australoids" (senso molto latto).

The lack of Paleolithic human remains though makes all this matter impossible to test anyhow. So we have to trust genetics.

There were flows through the North later on but they belong to a later moment: at least the beginnings of the UP.

terryt said...

"As to the main thesis of the paper, namely that East Asians are descended from Southeast Asians rather than Central Asians, I have to say that I am not convinced".

Nor am I. I think this sums the study up:

"Ten Asian countries came together in the spirit of solidarity to understand how we were related as a people, and we finished with a truly Asian scientific community".

They had a political objective.

"Genetic diversity increases in large populations".

Quite. We can be sure that any Paleolithic population in the north was very small. Life was hard. Unlike the south, relatively speaking.

"climatically and culturally (modes of sustenance), northern India/Pakistan was somewhat separated from the south - even in the absence of mountain ranges, by a wide swath of barren land until after LGM".

And the same 'wide swath of barren land' separated northern India/Pakistan from the Ganges Valley. So we have a problem with India being a straightforward route.

"I find it easier to get there from northern modern Afghanistan, Tajikistan, and Kyrgyzstan than via Northern Myanmar/ West China etc".

I agree, in spite of what Maju claims. Although 'West and East Eurasia are very much separated by deserts except at the narrow Altaian corridor' at such times the 'wide swath of barren land' in Northwest India would be just as effective a barrier. And 'Northern Myanmar/ West China' are densely forested and extremely mountainous. Not an easy route at all.

"Thereafter, one or several groups of people migrated North, mixed with previous settlers there"

What does that mean? who were these 'previous settlers'? Just the same people who had earlier moved north?

Maju said...

Not an easy route at all.

It's not any trekking contest, just a matter of population survival. Those areas provide resources and are passable by primitive barefooted humans (unlike the northern corridor that requires shoes and good quality clothes, only documented since early UP - the same as AMH presence in that area).

Objection rejected.

eurologist said...

I agree and disagree. As I stated previously, there is in fact evidence of a second, more southern route - but nearly not as southerly as the coast, and perhaps capable of forming a very important third population neglected in the paper discussed.

I will try to do a map early next year if I have time. The crux is whether the most northerly India was populated first East-to-West, or the other way around. There clearly were two distinct, successful modes of sustenance: the tropical forest and coastal fishing in a very, very narrow range away from the coast (except for Sahul) until after LGM , and then the moderate/cold river valleys and plains region with much more easily accessible grazing animals in the north from Afghanistan and south of the Himalayas to the intersection at NE India with ~China. Clearly, after adapting to the cold and plains big-game hunting (tents, smoking and cold/drying-preservation of meat, etc.), people in the latter region would be much more likely to have given rise to both Northern European populations (in the West) and NE Asians (in the East) than the tropical dwellers of a large size Sahul that then was uncovered by falling sea levels.

At any rate, there is plenty of evidence for a third population (the earliest non-negroid, straight hair, light skin, adapted for cold plains/valleys) in the northernmost parts of India and Asia in general.

Maju said...

Straight hair? How can you know how was hair variability some 60 or so millennia ago?

Unknown said...

All of the out of Africa population have almost straight hair.

I expect that our out of Africa ancestors were similar to the Australian aborigines and southern islanders. Dark brown-skinned, rather than black. Dark hair, slightly curly with a potential for blonde as a childhood attribute (later a selected neotency attribute in some populations).

I suspect we will find that very kinky hair and perhaps black skin will be more advanced African developments in the African cauldron of evolution.

Anonymous said...

Annie mouse : "I suspect we will find that very kinky hair and perhaps black skin will be more advanced African developments in the African cauldron of evolution."

Jarawas (women, in this case) from the Andaman island (Bay of Bengal), in the sea not too far from Myanmar, Y-DNA hg D* & mtDNA hg M2 and M4 :

http://www.survival-international.org/lib/img/gallery/Image_Galleries/jarawa/800x600/JARAWA30_medium.jpg

Maju said...

All of the out of Africa population have almost straight hair

Say what?!

The majority of Caucasoids do not have straight hair but curly or wavy. The native peoples south of SE Asia also don't typically have straight hair but curly, wavy or often even woolly in an African style. In the rest, the situaltion varies (you find pockets of curly hair in East Asia and South Asia, which we can consider "the flat hair regions").

terryt said...

"It's not any trekking contest, just a matter of population survival".

I'll repeat what I said at your blog regarding the situtaion in island SE Asia in Hoabinhian times.

http://epress.anu.edu.au/pima/pdf/ch06.pdf

Quote:

"An increasing density of vegetation in these areas would have affected hunting populations through a diminution in mammal biomass, which decreases dramatically as one moves from optimal savanna conditions, through parkland, towards rainforest".

That's what I keep trying to tell you, but you don't understand what it's actually like to hike through rainforest-covered hills.

The region of 'the Himalayas to the intersection at NE India with ~China' is exactly that sort of country. Difficult to either live in or pass through.

"The native peoples south of SE Asia also don't typically have straight hair but curly, wavy or often even woolly in an African style".

Quite. Straight hair is most common in the northern part of East Asia. And other East Asian physical characteristics, such as eye shape, are also more accentuated in the north. Is it really likely that the Northeast Asian phenotype evolved as the population moved north through the jungles of South China, to emerge pre-adapted to life on the open plains of Mongolia?

Maju said...

I'll repeat what I said at your blog...

And I'll quote what I wrote there as reply:

"So do you think this issues makes life impossible in the jungle or just less productive?

"People could easily live in such jungles and it was just a matter of time until some group adopted them as their hunting grounds and homeland. With the signs of exploding population we percieve in early Eurasian genetic data, this colonization happened surely sooner than later

(...)

"You develop too rigid and extreme conclusions from rather inconclusive comments, don't you?".

You also forget the coastal route, etc.

terryt said...

And my reply at your blog (rather long, I'm afraid).

Maju: "So do you think this issues makes life impossible in the jungle or just less productive?"

Me: Basically the only people who live in jungle today are those who've become phenotypically adapted to it (such as Pygmies and Negritos) or those who've set fire to it, cleared it and now farm it. It's not an environment people readily and willingly move into. Even the NZ Maori, to some extent a non-ceramic Neolithic people, did not inhabit the deep forest. They did occasionally set fire to it and then farm any more gently sloping denuded land.

Maju: "The damn genetic diversity issue! That's the evidence - and pretty much the only one available, AFAIK".

Me: Not the only evidence. Consider the giant panda's distribution:

http://nationalzoo.si.edu/Animals/GiantPandas/PandaHabitat/default.cfm

Me: And this:

http://mbe.oxfordjournals.org/cgi/content/full/24/8/1801

Quote from second link: "However, the giant panda had a much wider range in the Pleistocene, with fossil records from Zhoukoudian, near Beijing to southern China, and into northern Myanmar, northern Vietnam, Laos, and Thailand (Wen and He 1981; Zhu and Long 1983; Hu et al. 1985; Wei et al. 1990; Hu 2001)".

Another quote from second link: "Recent demographic inference using microsatellite markers demonstrated a clear genetic signature for population decline starting several thousands years ago or even futher back in the past, and being accelerated and enhanced by the expansion of human populations".

Me: Do you think the panda's survival in the mountains that you persistently claim people moved through so readily is because these early humans were considerate enough not to hunt it? I would guess the timing of its range contraction would tell us precisely when humans moved into its historical range.

Finally: Some may like to follow the complete discusssion:

http://leherensuge.blogspot.com/2009/11/genes-brains-and-fractals-various-brief.html

Maju said...

It's irrelevant to keep duplicating this discussion. The original one is at this link for whoever is interested.

ren said...

Eurologist, what makes you assume that "Caucasoids", or was it "West Eurasians" you said, are from NW India/Pakistan? If from Maju, he recycled it from me.

The point is that if you are getting a recycled idea, it's not worth much.

From the distribution of IJK, KT, and K, it rather looks like West Eurasians split off very early, east of Iran. There's no autochthonous IJ or even IJK or KT (the old K) in India so far found, so does that look like Europeans and Middle Easterners (autosomally and mtDNA-wise) be from South Asia or have something to do with Central Asia?

SetzerTeshub said...

Old news. I have laid out this exact East Asian population origin theory YEARS ago. You can find more here, including explanations for the laymen to understand.

http://s6.zetaboards.com/man/topic/8574252/?author=4012214

Out of Africa groups can be separated into 3 primordial sub-groups that resulted from being both isolated AND intermixing with older human subspecies. Your precious "Caucasoids" are only a younger branch of the overall West Eurasian trunk, and the 'purest' of them were from northwest Iran.

Besides them were other varieties of admixed Europeans who were part-neanderthal. Below them are the distant ancestors of Austromelanesians, inhabiting lands south of Mesopotamia as late as 10,000 years ago. These I call 'Indonoid' as its most fitting.

The crux is that pre-Caucasoid and Indonoids split much more closely, compared to pre-Mongoloids. From S. India to S.E. Asia, the ancestors of Mongoloids also intermixed with older humanoid species, giving them a different set of phenotypes compared to Westerners.

Sadly, we only have two hopelessly lost camps in this area. The Eurocentrists and Easterncentrists. The Euros put out an image of an advanced race of 'Whites' leaving their native European lands to conquer the south and set up civilizations. Their problems started when it become more than obvious that Scandinavia has significant northern Mongoloid contribution and the south is pretty much an extension of the Middle East. Uh oh. To solve this in the late 90's/early 2000's, they started to cal all Western (non-Mongoiloid) groups as 'European', but then reverted back to West Eurasian when their agenda started to break down again (this might be the second one after the defunct Caucasoid-Mongoloid split in Central Asia theory).

Now the Eastern or Sinocentrists like this ren person. Believed the East Asians (and most Japanese) are supremely unique beings who evolved in the north. And not only that, these early Mongoloids were the ones to who shared their 'seed' to made Europeans from neanderthals/austroloid pre-Caucasoid. Nice theory...except that Mongoloids were no were near western Mongolia even 12,000 years ago.

The final nail in the coffin for both came when it was finally revealed that C* and D* haplogroups have their origin in north India. So here we have the 'Indonoids', the missing piece that explains the origin of New Worlders, pre-Mongoloid Siberians, the Jomon, Ainu, and Austroloids, and many other Western groups that are otherwise just called "Caucasoid" for simplicity. Sorry ren, but you are a mongrel as well.

Maju said...

Stetzer: what you say is a total nonsense. You can't talk of "races" before these modern phenotypes coalesced (what only happened very slowly by regional convergence, not just global divergence).

Also, as anyone with a two fingers wide forehead can tell, haploid lineages are not "races". People of the same "race" can perfectly have almost totally unrelated lineages. Similarly early Eurasians, who were surely a single population initially, had at least two mtDNA lineages and at least two or three Y-DNA ones.

Ren: You talk of KT and K as if these would be already official. However up to today, ISOGG still has T as a subclade of K.

I presume that your "K", is what was recently described as MNOPS macro-haplogroup and that this is the same as previously detected as NOP. So basically it's K(xL,T), right?

Hence there is authoctonous IJK and K (your "KT") in India: it's called L and K1. It is not totally impossible that some of the ignored F-number lineages, that are mostly South Asian, also belong to IJK.

Whatever the case at the state of the art of Y-DNA knowledge F looks a lot like South Asian in origin, with 4/7 described subhaplogroups exclussive of the subcontinent. What IJK and K did after is more complicated to explain admittedly but, considering that IJ is West Eurasian and MNOPS looks as having a SE Asian homeland, I would not hurry in discarding a central role for South Asia in this secondary process anyhow. Y-DNA lineages obviously can either roll over or vanish too easily and the possible minor haplogroups in all these new matriushka subsets of F are still totally unresearched. Remember that before the "F*" diversity was somehow understood almost nobody thought this lineage had a South Asian origin, but a Western one. The same is surely the case with all these new sublineages: we only see the big stuff and we must be missing crucial details in the "asterisk" zone.

terryt said...

"MNOPS looks as having a SE Asian homeland"

As most of you are well aware I've long argued for a specifically Wallacean origin for these SE Asians. But I've now expanded the region. I think it included what was, at times of lowered sea level, the almost landlocked South China Sea as well.

And I wouldn't dismiss the Malay Peninsular as being of marginal importance for SE Asian humans. Rather than being a sidetrack on any overland journey into Sunda it would, in fact, be the main highway.

Sea level needs to fall only slightly before the peninsular becomes connected to Sumatra. A further drop simply connects the Malay Peninsular to Borneo as well as Sumatra. In fact I suspect that this latter was actually the main route into Sundaland. The eastern side of Sumatra is basically extensive flood plain jungle, formed from debris off the rapidly rising western side of the island.