August 05, 2012

1000 Genomes Project Community meeting video

can be found here.

I fast forwarded through a few of the talks. Some interesting tidbits:

1) John Novembre reports on the human autosomal mutation rate on the basis of a very large sample sequenced in a small number of regions; he is using an approach that simultaneously estimates effective size and mutation rate: he gets a median of 1.38x10^-8 per bp per gen. This tends to agree with previously published pedigree estimates, and seems to also be quite lower than a widely used value of 2.5x10^-8 that assumed an ancestral effective population size and an age for the human-chimp divergence. Adoption of the slower rates has implications: if people start using the lower rates that come out of sequencing, population splits will probably have to be redated.

2) Jay Shendure speaks about the future of sequencing; we are now at around $3,500; he expects costs to continue to decline, but we shouldn't be overly optmistic due to technology limits+market forces. I think that the latter may be significant: when a company can offer a better deal than any of its competitors, it has no incentive to drop prices further, even if its own costs are dropping; moreover, demand is going to surge as prices approach the magical $1,000 mark: I know I'll be very tempted to buy at that point myself.

3) Alexander Platt talks about inferring population history using haplotypes. Of interest: during the last 1,000 generations there are more coalescences between Beijing Chinese and Japanese rather than Beijing Chinese and southern Chinese; in more recent times, there are more coalescences between Chinese groups. This makes some sense, if we suppose that -as seems likely- Mongoloids spread north-to-south across China during prehistory; the Japanese are thus linked -in older times- with northern Chinese, both of which are mostly descended from the northern Mongoloids; in more recent times, especially after the emergence of a uniquely Chinese polity and culture, the Chinese tend to marry other Chinese, hence they share more recent common ancestors within the country itself.

15 comments:

Matt said...

Possibly I'm misunderstanding here but:

Isn't 1000 generations like 20000 - 30000 years?

So equal coalescence between the Vietnamese, Beijing Chinese, Southern Chinese and Japanese groups would prior to 20000 to 30000 years.

Presumably equal coalescence would indicate that these populations are all living as one unstructured population (there are other scenarios where there would be no differences in how much they breed with one another - e.g. being already equally differentiated populations who are not breeding differentially).

Does that match up with what we thought about the population history?

Maju said...

... "the slower rates has implications"...

Indeed: everything gets twice older, everything else equal.

T. Kosmatka said...

"Indeed: everything gets twice older,"

Those early fossil dates in Australia make more sense then.

terryt said...

"Of interest: during the last 1,000 generations there are more coalescences between Beijing Chinese and Japanese rather than Beijing Chinese and southern Chinese; in more recent times, there are more coalescences between Chinese groups. This makes some sense, if we suppose that -as seems likely- Mongoloids spread north-to-south across China during prehistory"

But many still seem committed to denying any such north to south movement. It has been obvious for many years that the early SE Asian population was more similar to modern Papuans/Australians than to Mongoloids. The modern SE Asian are the product of a Mongoloid phenotype overlying that Papuan phenotype. Remnants of it remain in parts of SE Asia, notably Timor. The most likely candidate for having transported that phenotype south is Y-DNA O, along with several mt-DNA haplogroups. Likely candidates include M7, M8, M9, G, A and N9/Y.

Lathdrinor said...

@terryt the problem with the O expansion hypothesis for Mongoloid phenotypes is that O reaches its greatest diversity within SE Asia itself. Several subclades of O, believed to be primary branches, are not even found further north. Under the model in which focal of expansion ~ area of greatest diversity, O is therefore believed to have expanded from the south.

But that, of course, does not prevent undifferentiated O from having arrived in the south from the north. The challenge is to show that O populations were the carriers of Mongoloid phenotypes, which thus far has not been done. I do find it telling that all modern populations with haplotype O tend to carry a series of Mongoloid phenotypes - though this is neither complete nor uniform - but I also have to draw attention to the prehistoric link between Native American populations and Mongoloid populations, with the former carrying no O whatsoever.

I feel there is a missing link here that we have yet to discover.

terryt said...

"the problem with the O expansion hypothesis for Mongoloid phenotypes is that O reaches its greatest diversity within SE Asia itself".

But, as Dienekes often points out, diversity does not equal origin. Diversity can be a product of any number of factors. In this case the diversity is most likely a product of the convergence of several different migration routes on SE Asia. The three O haplogroups look to have each used a different route into that region. I am reasonably convinced that O's expansion is related to the Chinese Neolithic.

"Several subclades of O, believed to be primary branches, are not even found further north".

Such as? I realise O1 is not found very far north but it could still have originated in the central Yangtze region before it moved south overland to Taiwan and SE Asia. Its presence in SE Asia is associated with the Austronesian expansion. O2a is largely SE Asian but is probably associated with the expansion of Austro-Asiatic, presumably a Neolithic expansion, and its close relation O2b is primarily Korean. O2's movement south looks likely to have been coastal. As far as I'm aware no basal O3 haplogroups are found in the south that cannot be connected to a Neolithic culture. Its movement looks to have been overland from somewhere round the China/Tibet border, or the Upper Yangtze.

"But that, of course, does not prevent undifferentiated O from having arrived in the south from the north".

To me it looks most likely that differentiated O moved south. As for a southern origin for O, we have the problem that it appears to diversify almost as soon as it splits from N. And N is a northern haplogroup. My guess is that the phenotype developed in the relatively high altitude Mongolia/China/Tibet region, and was carried outwards from there.

"The challenge is to show that O populations were the carriers of Mongoloid phenotypes, which thus far has not been done".

But it is very difficult to come up with any other candidate as having spread the phenotype.

"I do find it telling that all modern populations with haplotype O tend to carry a series of Mongoloid phenotypes - though this is neither complete nor uniform"

Exactly. Although I disagree with the claim it the connection is not complete. As far as I'm aware all populations with O haplogroups have at least a reasonable element of Mongoloid phenotype. And the probability that the phenotype originated somewhere north of SE Asia is, to me, the only reasonable assumption.

"I also have to draw attention to the prehistoric link between Native American populations and Mongoloid populations, with the former carrying no O whatsoever".

No, but they do have C3, another carrier of the Mongoloid phenotype. And many would claim that Native Americans carry something other than a Mongoloid phenotype as well. Their migration may have been led by Y-DNA Q, probably Central Asian or Iranian in origin. The route also may have been along the northern margin of where the Mongoloid type had reached at that stage, because what became the Native Americans did pick up mt-DNA that seems associated with Mongoloid phenotype. Hence the apparent mixed phenotype of Native Americans.

Maju said...

@Terry: wishful thinking. Well understood diversity, specially basal diversity, means likely origin. We can't control for every single detail of past prehistory but we can't either dismiss maximum likelihood as the result of "any number of factors".

Also, let's be clear: the Mongoloid phenotype(s) is/are not only connected with Y-DNA O or its parent NO but also with many other largely unrelated lineages like C3, all D or Q1a. Assuming that phenotypes are tightly related with Y-DNA lineages is in itself a waste of time (look at the Finnish for example).

First of all we should discern IF there is in fact a single Mongoloid phenotype and is not an artifact or our minds. Dienekes himself in a previous article of anthropometric self-research, found that "Mongoloid" skulls split in several groups (one of them indeed akin to the areas dominated by Y-DNA O and D, which I dub "Sinoid", but not the C3 and Q1a areas of Mongolia, Siberia and Native America, which had their own distinct clusters), groups that had little to no relation with each other.

But whatever the case, phenotype, if anything must be related with autosomal DNA, not Y-DNA lineages, which are only obliquely related.

Phenotypes apart, the following Y-DNA haplogroups show a clear South to North pattern of expansion (each determined in one or several previous studies):

1. D
2. C (although C3 coalesced probably in NE Asia itself)
3. NO - and within it N and O - and within O at least O1 and O3

Those are the genetic facts and the anthropometric facts as far as I can discern. There is absolutely no basis to keep waving the old obsolete hypothesis of North-to-South main migration, regardless of less important secondary flows in that sense.

terryt said...

"But whatever the case, phenotype, if anything must be related with autosomal DNA, not Y-DNA lineages, which are only obliquely related".

But it is surely reasonable to suppose that the phenotype did not move south independent of any male or female movement.

"Also, let's be clear: the Mongoloid phenotype(s) is/are not only connected with Y-DNA O or its parent NO but also with many other largely unrelated lineages like C3, all D or Q1a".

C3, D and Q are all rare in SE Asia. And the phenotype is most noticeable where O is common. In fact in most cases directly proportional.

"First of all we should discern IF there is in fact a single Mongoloid phenotype and is not an artifact or our minds. Dienekes himself in a previous article of anthropometric self-research, found that 'Mongoloid' skulls split in several groups"

No doubt there is more than one gene involved. And differences in expression.

"but not the C3 and Q1a areas of Mongolia, Siberia and Native America, which had their own distinct clusters)"

And clearly distinct from SE Asian haplogroups.

"Phenotypes apart, the following Y-DNA haplogroups show a clear South to North pattern of expansion (each determined in one or several previous studies):
1. D
2. C (although C3 coalesced probably in NE Asia itself)
3. NO - and within it N and O - and within O at least O1 and O3"

You can hardly claim 'a clear South to North pattern' for any of those haplogroups. At most a 'possible south to north pattern'. Seeing the south to north pattern relies completely on accepting the great southern coastal migration as proven. Other interpretations are possible for the expansion of those haplogroups. For clarification we need further information. The direction of D's expansion is far from clear. A south to north pattern is certainly discernible if we're prepare to ignore several minor haplogroups. As you say C3 almost certainly coalesced in NE Asia but we have no way of discerning its route to there. N certainly doesn't demonstrate a south to north pattern as it's almost unknown in the south. If anything O1 can only have moved south into SE Asia, and much the same holds for O3.

Maju said...

@Terry:

Inuits have no O (Q1a instead) and are "Mongoloid" by all classical standards, Mongols are the very archetype and are dominated by C3 or N, Japanese and Tibetan are archetypally "Mongoloid" and half their lineages are D.

I'm not even looking to SE Asia yet. The commonality of Y-DNA O stops at around Manchuria but the "Mongoloid" phenotype is dominant much further North (and to the Northwest and the Northeast into America).

"You can hardly claim 'a clear South to North pattern' for any of those haplogroups".

I do, there is solid literature on the matter in all cases (which you know about) but (AFAIK) C but is so obviously more diverse in SE-Asia-cum-Australasia that it's not even needed to debate, as long as you accept that a haplogroup's basal diversity is relevant.

Not accepting that either forces you to make a point argument for each haplogroup and discuss what evidence you'd think that would justify the exception (casuistic), or makes you a pseudoscientist for which there is no method but a pre-conceived model to defend no matter what.

I'm not interested in the latter, but if you'd have a casuistic approach with actual evidence for each exception claimed (what I know from previous discussions that you do not), I'd like to know about it.

terryt said...

"Inuits have no O (Q1a instead) and are 'Mongoloid' by all classical standards"

So?

"Mongols are the very archetype and are dominated by C3 or N, Japanese and Tibetan are archetypally 'Mongoloid' and half their lineages are D".

Very true. But what is the significance of that in relation to O? Doesn't it imply that O originated somewhere near haplogroup C3? It certainly must have originated somewhere near haplogroup N. And quite probably all those 'Mongoloid' haplogroups came in contact with D at some early stage.

"I'm not even looking to SE Asia yet. The commonality of Y-DNA O stops at around Manchuria but the 'Mongoloid' phenotype is dominant much further North"

And exactly what do you see as the problem in explaining that situation? I certainly can see no problem. Anyway your above comments are a bit rich, coming from someone who just yesterday wrote:

"Assuming that phenotypes are tightly related with Y-DNA lineages is in itself a waste of time"

And here you go assuming exactly that.

"I do, there is solid literature on the matter in all cases (which you know about) but (AFAIK) C but is so obviously more diverse in SE-Asia-cum-Australasia that it's not even needed to debate"

Haplogroup C is irrelevant here. What we're concentrating on is Mongoloid haplogroups. C3 is most definitely not most diverse in SE Asia. You can hardly claim a south to north expansion for the Mongoloid haplogroup C3 through China.

"as long as you accept that a haplogroup's basal diversity is relevant".

I keep trying to explain to you that it is not always relevant. Other factors come into play.

"but if you'd have a casuistic approach with actual evidence for each exception claimed (what I know from previous discussions that you do not), I'd like to know about it".

OK. How would you explain the spread into SE Asia of the Mongoloid phenotype? The only explanation that makes any sense to me is that it was carried there by men and women. These men and women must have carried haplotypes. These haplotypes should remain in SE Asia somewhere.

Actually I know you well enough to know already what you're going to claim. You're going to say that the Mongoloid phenotype developed over the whole of East Asia without any human movement whatsoever. The basis of your claim seems rather racist to me. You are prepared to admit that human movement since the Upper Paleolithic has given rise to the distribution of haplogroups in Europe but you maintain that there has been no similar movement in East or Southeast Asia since modern humans first arrived there. An extremely unlikely scenario.

Maju said...

"So?"

So there are many "Mongoloids" who bear no or trivial relation with Y-DNA O. Y-DNA O is part of the "Mongoloid" landscape but not its origin nor its vector. The phenotype is only related to autosomal DNA and in this case it corresponds with the wider East Asian genetic cluster essentially.

And in that area of East Asia all haplogroups at deep level look like migrating from South to North, which is consistent with the generally accepted model for the original colonization of East Asia some 60 or 70 Ka ago.

So that.

terryt said...

"So there are many 'Mongoloids' who bear no or trivial relation with Y-DNA O".

You're making the same mistake as you consistently make concerning the Austronesians. The Austronesians contained several different haplogroups as do the Mogoloids. I would have thought it was simple to understand, but obviously not. Y-DNA O is closely associated with the Mongoloid phenotype in SE Asia regardless of the relationship between the two elsewhere.

"Y-DNA O is part of the 'Mongoloid' landscape but not its origin nor its vector".

Y-DNA O is the Mongoloid phenotype's only possible vector as far as SE Asia is concerned. Of course there is minimal C3 in SE Asia but not enough to explain the considerable level of the Mongoloid phenotype there. And presumably some mt-DNA haplogroups were also involved in the spread.

"The phenotype is only related to autosomal DNA and in this case it corresponds with the wider East Asian genetic cluster essentially".

But it cannot have instantly appeared through such a widespread region. It must have been carried by men and women. I do agree on the other hand that autosomal DNA can spread beyond a haplogroup's spread. But usually autosomal DNA's spread will start with the spread of humans with a particular haplogroup(s).

"And in that area of East Asia all haplogroups at deep level look like migrating from South to North, which is consistent with the generally accepted model for the original colonization of East Asia some 60 or 70 Ka ago".

Quite possibly so. But in no way can the Mongoloid phenotype have spread in that direction. It can only have spread from north to south into SE Asia, and probably even southern China. Presumably NO was the haplogroup that moved from south to north. It presumably spread from within the region that haplogroup MNOPS had managed to occupy. And probably from the northern margin of that spread.

Maju said...

Last comment on the "Mongoloid" theme because we can go forever:

"Y-DNA O is the Mongoloid phenotype's only possible vector as far as SE Asia is concerned".

Only IF you believe that the phenotype arrived there from the North as some sort of demic wave. And, even in that case, it does not match because the O haplogroup has all the indicators of having expanded in general and in most parts from South to North.

"But it cannot have instantly appeared through such a widespread region".

They are not clones: the only coherence of the morphotype is the same as the coherence of the autosomal genotype. And that looks mostly deeply rooted in the Paleolithic if we are to judge from Fst distances and such.

"But in no way can the Mongoloid phenotype have spread in that direction. It can only have spread from north to south into SE Asia"...

That's your faith. I insist anyhow: there is not any single one Mongoloid phenotype just a general areal affinity like between Sweden and Iraq. In the West nobody questions anymore that all populations originated ultimately in West Asia. Why? Because genetics indicates so.

Same in East Asia.

Said that, there was probably a "Mongoloid-cum-O" migration into Sundaland. We don't know exactly when but, judging from the Karafet 2010 paper, it should also be very old. This can probably be compared to the West Eurasian flows into Arabia and North and East Africa.

An this is the last I'll say on this matter.

terryt said...

"Only IF you believe that the phenotype arrived there from the North as some sort of demic wave".

What other possibility is there?

"Said that, there was probably a 'Mongoloid-cum-O' migration into Sundaland. We don't know exactly when but, judging from the Karafet 2010 paper, it should also be very old".

Thank you. You've just contradicted yourself. You suddenly admit there was some sort of Mongoloid migration into SE Asia. But, in spite of what the Karafet paper tried to prove, it is much more recent than the Paleolithic. I seem to remember offering multiple reasons for the faulty reasoning in that paper on your blog at the time. But you were just as committed to your faith then as you still are.

"the O haplogroup has all the indicators of having expanded in general and in most parts from South to North".

That statement is complete rubbish. Have another look at the distribution of the O haplogroups.

"the only coherence of the morphotype is the same as the coherence of the autosomal genotype. And that looks mostly deeply rooted in the Paleolithic if we are to judge from Fst distances and such".

No. Most see the SE Asian human remains from the Hoabinhian period as being 'Papuan' in phenotype, not 'Mongoloid'. Therefore the Mongoloid phenotype is almost certainly more recent in SE Asian than the Hoabinhian period.

"I insist anyhow: there is not any single one Mongoloid phenotype just a general areal affinity like between Sweden and Iraq".

The Mongoloid phenotype forms a cline of decreasing expression all the way from North of China down into island SE Asia. That cline even shows up within China, with south Chinese generally being less Mongoloid than north Chinese. The Mongoloid phenotype failed to reach New Guinea (until very recently) or Australia and only marginally reached Timor. However it does spread out into the Pacific islands. Those particular islands have only been settled in the last 5000 years. Their settlement was probably probably part of an continuous and ongoing expansion of the phenotype.

"Said that, there was probably a 'Mongoloid-cum-O' migration into Sundaland. We don't know exactly when but, judging from the Karafet 2010 paper, it should also be very old".

Thank you. You now admit there was some sort of Mongoloid migration into SE Asia. But, in spite of what the Karafet paper tried to prove, it is much more recent than the Paleolithic. I seem to remember offering multiple reasons for the faulty reasoning in that paper on your blog at the time. But you were just as committed to your faith then as you still are.

"In the West nobody questions anymore that all populations originated ultimately in West Asia. Why? Because genetics indicates so".

And in this part of the world nobody questions that the SE Asian population is the product of a Mongoloid phenotype overlying a Papuan phenotype. The genetics indicates so, if interpreted with an open mind rather than a mind that has decided in advance what it wants to see.

terryt said...

"An this is the last I'll say on this matter".

But I have more to add, for Maju's benefit obviously. I realise he will stick to his preconceived notion no matter how much evidence is presented though. This is the best examination of the Hoabinhian phenotype I've been able to find:

http://s1.zetaboards.com/anthroscape/topic/4646060/1/

Quote:

"According to this paper, the Hoabinhian tool industry in Vietnam was associated with Australoids".

And:

"In general terms, Southeast Asia is thought to have been originally occupied by indigenous people (sometimes referred to as Australo-Melanesians) that subsequently exchanged genes with immigrants from North and/or East Asia, during the Holocene, leading to the formation of present-day Southeast Asians (Callenfels, 1936; Mijsberg, 1940; Von Koenigswald, 1952; Coon, 1962; Jacob, 1967). More recent studies based on late Pleistocene and early Holocene human remains represented by specimens from Niah Cave in Borneo (Brothwell, 1960; Kennedy, 1977; Barker et al., 2007), Tabon Cave on Palawan Island, Philippines (Fox, 1970; Macintosh, 1978; Dizon et al., 2002), Gua Gunung Runtuh in Peninsular Malaysia (Zuraina, 1994, 2005; Matsumura and Zuraina, 1999) and Moh Khiew Cave in Thailand (Matsumura and Pookajorn, 2005) have provided additional support for the existence of an ‘Australo-Melanesian’ lineage in ancient Southeast Asia"

Note: 'Holocene', not earlier. A couple of other interesting points that are probably relevant:

The survival until relatively recently of orangutans and pandas in South China suggests the region was sparsely settled until quite recently.

Haplogroup N appears mainly found on the plateau region north of Northern China and further north. Haplogroup O tends to be concentrated south of that region. To me it looks likely that the two haplogroups parted company somewhere in the region on the boundary between the two.