June 05, 2008

500K SNP study of Oceanian populations

Related: Genetic structure of Pacific Islanders

Molecular Biology and Evolution, doi:10.1093/molbev/msn128

Gene Flow and Natural Selection in Oceanic Human Populations, Inferred from Genome-wide SNP Typing

Ryosuke Kimura et al.

It is suggested that the major prehistoric human colonizations of Oceania occurred twice, namely, about 50,000 and 4,000 years ago. The first settlers are considered as ancestors of indigenous people in New Guinea and Australia. The second settlers are Austronesian-speaking people who dispersed by voyaging in the Pacific Ocean. In this study, we performed genome-wide SNP typing on an indigenous Melanesian (Papuan) population, Gidra, and a Polynesian population, Tongans, by using the Affymetrix 500K assay. The SNP data were analyzed together with the data of the HapMap samples provided by Affymetrix. In agreement with previous studies, our phylogenetic analysis indicated that indigenous Melanesians are genetically closer to Asians than to Africans and European Americans. Population structure analyses revealed that the Tongan population is genetically originated from Asians at 70% and indigenous Melanesians at 30%, which thus supports the so-called "Slow train" model. We also applied the SNP data to genome-wide scans for positive selection by examining haplotypic variation, and identified many candidates of locally selected genes. Providing a clue to understand human adaptation to environments, our approach based on evolutionary genetics must contribute to revealing unknown gene functions as well as functional differences between alleles. Conversely, this approach can also shed some light onto the invisible phenotypic differences between populations.

Link

11 comments:

terryt said...

"indigenous Melanesians are genetically closer to Asians than to Africans and European Americans". They don't specify whether East Asians or South Asians. But Maju, do you remember asking me about a funny map in my "Into Australia" essay? Seems this study is support for Cavalli-Sforza's principal complonent map.

Polak said...

Closer to East Asians - Chinese and Japanese samples from the HapMap project.

Richard Parker said...

Could someone please translate this for me?

"We also applied the SNP data to genome-wide scans for positive selection by examining haplotypic variation, and identified many candidates of locally selected genes. Providing a clue to understand human adaptation to environments, our approach based on evolutionary genetics must contribute to revealing unknown gene functions as well as functional differences between alleles. Conversely, this approach can also shed some light onto the invisible phenotypic differences between populations."

What, exactly, are 'invisible phenotypic differences' ?

It would be very exciting if "indigenous Melanesians WEREN'T genetically closer to Asians than to Africans and European Americans"

This team has squeezing the the last juice out of just two populations, Gidra and Tongans for at least two years now:

http://www3.interscience.wiley.com/journal/112265847/abstract?CRETRY=1&SRETRY=0

in which they suggested:

"that the expansion of Polynesian ancestors possessing these variations occurred approximately 7,000 years ago" - possibly right, but purely by accident.

The Gidra come from the Fly River area in South PNG, about as pure New Guinean as you can get. BUT, they've been visited annually by the Austronesian-speaking Motu from around Port Moresby on their annual Hiri voyages, trading pots for sago, and everyone knows what sailors do when they eventually get to (any) port.

Dienekes - you really shouldn't waste your (very good) blogging time on this kind of crap.

Unknown said...

Data mining and re-publishing with new analysis happening. People need to publish their papers I guess.

The previous frequently used words were

South Asian( Indian) Tribes are considered genetically closer to Asians. This was due to common paleolithic antiquity and shared mt Haplogroup M and derivatives.

I am not sure anybody measured neighbor joining tree distance for Indian Tribes and indigenous Melanesian (Papuan) populations.

Both are C+(M,N) in its form.

terryt said...

South central haplo. Melanesian males are more often Y-chromosome K and its descendant M rather than C.

Richard. Austronesians and Melanesians have been mixing along the coastlines since austronesians first arrived there.

Unknown said...

Terry than what is your basis for Y haplo C spread from Melanesia argument

terryt said...

South central haplo. Polynesian C y-chromosome is related to Melanesian C Y-chromosome (C2 if I remember correctly). This particular Y-chromosome is also found in Eastern Indonesia and may have been primarily coastal. It is certainly more southern in origin than the Polynesian mtDNA, primarily B, so it seems it was a hybrid population who moved into the wider Pacific.

Richard Parker said...

Terryt - certainly Austronesians and NANs met and bred in Island Melanesia (New Britain and New Ireland) and linguists (Pawley & Ross) will have it that they founded proto-Oceanic, and spread out from there, to go on to glory in Polynesia.

However, there's a very strong dividing line between the Austronesian Meso-Melanesian (Island) languages, and the much older North New Guinea Cluster (Mainland), that meet in New Britain.

Meso-Melanesian has 35-40% cognates with 'proto-Oceanic' and the North New Guinea Cluster often has less than 10% - suggesting they've had a lot more time to change.

Meso-Melanesian (Island) languages segue straight into Polynesian, but the North New Guinea (Mainland) ones don't. They do, however, with Vanuatu.

There are no Austronesian speakers at all on the South Coast between about the longitude of Port Moresby and the Bird's Head.

That's between the Bird's a*hole and it's neck.

The Gidra, that these guys are referring to, are about 1/3 the way along.

terryt said...

"there's a very strong dividing line between the Austronesian Meso-Melanesian (Island) languages, and the much older North New Guinea Cluster (Mainland) ... the North New Guinea Cluster often has less than 10% - suggesting they've had a lot more time to change". More likely the 10% are borrowings. The two language groups seem not to be even distantly related. This probably indicates two separate movements through the region. But languages can be independent of Y-chromosomes or even mtDNA. Even historically many groups have borrowed languages that have only a recent connection with them.

"There are no Austronesian speakers at all on the South Coast". Which supports other evidence that the Austronesian languages moved along the north coast.

I wasn't aware of the connection between Vanuatuan languages and any mainland New Guinea ones. Do you have a link?

terryt said...

As usual it wasn't until today I realised where the confusion arose. Micronesian Languages are divided into two: west (Yap, Pelau etc.) and east (Carolines, Kiribati etc). The western ones do seem to be related to some members of the New Guinea/Melanesian complex. But the eastern ones, as you say, are related to Vanuatuan languages. They are Austronesian but are distinct from the Polynesian languages from even further east.

As far as I'm aware the Melanesian languages do not extend into the Pacific beyond Western Micronesia and the Northern Solomons. Beyond the Northern Solomons the languages are all Austronesian (apart from those introduced since European arrival).

terryt said...

I have now seen that the west Micronesian languages are also Austronesian. Even in the Solomon Islands only a very few of the northern islands actually have Melanesian or non-Austronesian languages.