April 05, 2008

Austronesian expansion in Admiralty Islands of Melanesia

Molecular Biology and Evolution, doi:10.1093/molbev/msn078

The impact of the Austronesian expansion: evidence from mtDNA and Y-chromosome diversity in the Admiralty Islands of Melanesia

Manfred Kayser et al.

The genetic ancestry of Polynesians can be traced to both Asia and Melanesia, which presumably reflects admixture occurring between incoming Austronesians and resident non-Austronesians in Melanesia before the subsequent occupation of the greater Pacific; however, the genetic impact of the Austronesian expansion to Melanesia remains largely unknown. We therefore studied the diversity of non-recombining Y-chromosomal (NRY) and mitochondrial (mt) DNA in the Admiralty Islands, located north of mainland Papua New Guinea, and updated our previous data from Asia, Melanesia and Polynesia with new NRY markers. The Admiralties are occupied today solely by Austronesian-speaking groups, but their human settlement history goes back 20,000 years prior to the arrival of Austronesians about 3,400 years ago. On the Admiralties we found substantial mtDNA and NRY variation of both Austronesian and non-Austronesian origin, with higher frequencies of Asian mtDNA and Melanesian NRY haplogroups, similar to previous findings in Polynesia, and perhaps as consequence of Austronesian matrilocality. Thus, the Austronesian language replacement on the Admiralties (and elsewhere in Island Melanesia and coastal New Guinea) was accompanied by an incomplete genetic replacement that is more associated with mtDNA than with NRY diversity. These results provide further support for the "Slow Boat" model of Polynesian origins, according to which Polynesian ancestors originated from East Asia but genetically mixed with Melanesians before colonizing the Pacific. We also observed that non-Austronesian groups of coastal New Guinea and Island Melanesia had significantly higher frequencies of Asian mtDNA haplogroups than of Asian NRY haplogroups, suggesting sex-biased admixture perhaps as a consequence of non-Austronesian patrilocality. We additionally found that the predominant NRY haplogroup of Asian origin in the Admiralties (O-M110) likely originated in Taiwan, thus providing the first direct Y-chromosome evidence for a Taiwanese origin of the Austronesian expansion. Furthermore, we identified a NRY haplogroup (K-P79, also found on the Admiralties) in Polynesians that most likely arose in the Bismarck Archipelago, providing the first direct link between northern Island Melanesia and Polynesia. These results significantly advance our understanding of the impact of the Austronesian expansion and of human history in the Pacific region.



Ebizur said...

I can't believe how much crap some geneticists expect people to swallow. If you are getting results that suggest to you that there has been an invasion of only females or that has made a larger impact on the mtDNA than on the Y-DNA of the colonized region, then your assumptions of the affiliation of some mtDNA or Y-DNA haplogroup are incorrect. Stop trying to convince people that it is otherwise. It is obvious to any human being with half a brain and real-world experience that the influence of an invading group of foreign origin should have an equal or greater effect on the Y-DNA than on the mtDNA. I presume in this case that the researchers refuse to re-evaluate their assumption of an indigenous Australo-Melanesian, non-Austronesian origin of haplogroup C2-M38. Haplogroup C2-M38 is found in as much as 50% of the male population of some islands of central Indonesia, besides being the Y-chromosome haplogroup of the majority of Polynesian males, who have recently been shown to be autosomally East Asian; I strongly doubt that haplogroup C2-M38 is originally of Papuan origin.

Ebizur said...

The key points here are:

*The Admiralty Islands are occupied today solely by Austronesian-speaking groups

*The present inhabitants of the Admiralty Islands, who speak only Austronesian languages, are claimed to have a significant amount of Austronesian mtDNA but little or no Austronesian Y-DNA (This should raise a red flag!)

*Some Central Indonesians (e.g. Sumbanese, who are, like the Admiralty Islanders, Austronesian speakers) have very high frequencies of haplogroup C2-M8, the haplogroup which these researchers claim to represent an indigenous Papuan paternal heritage of the Austronesian-speaking inhabitants of the Admiralty Islands

*Papuan speakers do NOT have a high frequency of haplogroup C2-M8; the frequency of this haplogroup among inhabitants of Papua New Guinea is highest on average among Austronesian-speaking coastal groups, who also tend to have high frequencies of haplogroup M, whereas the highland Papuans are distinguishable by their having significant frequencies of haplogroup S. Haplogroup C2 is most frequent in Polynesia and central-eastern Indonesia (Sulawesi, Nusa Tenggara, and the Malukus), while being much less frequent in New Guinea, although still having some presence among populations of coastal New Guinea.

*Haplogroup M is the typical Y-chromosome haplogroup of Papuan-speaking inhabitants of islands of Melanesia that have Papuan speakers, which suggests that haplogroup M was carried by some non-Austronesian-speaking group that was involved in close interactions with Austronesian-speaking groups, while the haplogroup S carriers tended to stay isolated in the interior of New Guinea

*Autosomal studies have found inhabitants of coastal Papua New Guinea to be much more closely related to Melanesian islanders than to the inhabitants of the mountainous interior of Papua New Guinea, who are autosomally more similar to Australian aborigines

Try again, Kayser et al.!

Ebizur said...

If the Y-DNA that the authors are claiming to represent Papuan lineage is haplogroup C2, then I won't buy it so easily. If they are claiming that the Admiralty Islanders have a high frequency of haplogroup M, on the other hand, then their conclusions might be more believable.

Here is the distribution of NRY variation among some samples of East Indonesians, Australian aborigines, Papua New Guineans, Melanesians, Micronesians, and Polynesians in Hammer et al. (2005):

East Indonesia (n=55):
C*-RPS4Y(xM217, M38, M8) = 5.5%
C2*-M38(xP33) = 27.3%
K*-M9(xM230, M20, M5, M214, LLY22g, M175, P27) = 20.0%
S-M230 = 12.7%
M1-M5 = 12.7%
O3*-M122(xM134, LINE1) = 9.1%
O1a*-M119(xM110) = 9.1%
O1a2-M110 = 3.6%

Australian aborigines (n=33):
C*-RPS4Y(xM217, M38, M8) = 48.5%
K*-M9(xM230, M20, M5, M214, LLY22g, M175, P27) = 42.4%
R-M207 = 9.1%

Papua New Guinea (n=46):
C*-RPS4Y(xM217, M38, M8) = 2.2%
C2*-M38(xP33) = 8.7%
K*-M9(xM230, M20, M5, M214, LLY22g, M175, P27) = 54.3%
M1-M5 = 28.3%
O3*-M122(xM134, LINE1) = 4.3%
O1a*-M119(xM110) = 2.2%

Melanesia (n=53):
C2*-M38(xP33) = 18.9%
J-12f2 = 1.9%
K*-M9(xM230, M20, M5, M214, LLY22g, M175, P27) = 24.5%
S-M230 = 7.5%
M1-M5 = 35.8%
O3*-M122(xM134, LINE1) = 5.7%
O1a*-M119(xM110) = 1.9%
O1a2-M110 = 1.9%
R-M207 = 1.9%

Micronesia (n=17):
C*-RPS4Y(xM217, M38, M8) = 5.9%
C2*-M38(xP33) = 5.9%
D2a*-M116(xM125) = 5.9%
K*-M9(xM230, M20, M5, M214, LLY22g, M175, P27) = 35.3%
M1-M5 = 5.9%
O3*-M122(xM134, LINE1) = 17.6%
O1a*-M119(xM110) = 11.8%
O2b1*-P49(x47z) = 5.9%
P*-P27(xP36, M207) = 5.9%

Polynesia (n=60):
C2*-M38(xP33) = 3.3%
C2a-P33 = 60.0%
I-P19 = 1.7%
K*-M9(xM230, M20, M5, M214, LLY22g, M175, P27) = 3.3%
M1-M5 = 3.3%
O3*-M122(xM134, LINE1) = 18.3%
O3a5-M134 = 5.0%
O3-LINE1 = 1.7%
O1a*-M119(xM110) = 3.3%

The "Papua New Guinea" sample of Hammer et al. must have been taken from a population of the coastal region that lacks S-M230 and instead has a high frequency of M1-M5, which is also suggested by the presence of haplogroup O3* (2/46) and O1a* (1/46) in this sample, as haplogroup O-M175 in New Guinea is almost exclusively found among coastal populations that have a high frequency of haplogroup M. Note that Papua New Guinea, which is heavily Papuan-speaking, has a low frequency of haplogroup C2* (only 8.7% in this obviously Coastal PNG sample), whereas Melanesia, which is heavily Austronesian-speaking, has a relatively high frequency of haplogroup C2* (18.9%), and the peak frequency of haplogroup C2* is found in the East Indonesian sample (27.3%). Eastern Indonesia is, like Melanesia, also a heavily Austronesian-speaking region.

terryt said...

OK Ebizur. The main point is that people on the Admiralty Islands appear to be a mix of people. The female line comes mainly from Taiwan and the male line primarily from south of that, ialsnd SE Asia or whatever.

You wrote, "If you are getting results that suggest to you that there has been an invasion of only females". The evidence suggests the offshore islands were uninhabited and a mix of people were the first to reach them.

Dragon Horse said...

There didn't have to be an "invasion" of females. It could have been the women were brought there as slaves. It could also been that the original population was small and a group of warrior males came and basically took over...(think Latin America, or possibly England) causing a reduction in fertility of the aboriginal male population.

Ebizur said...

If the supposedly Asian mtDNA-bearing, Austronesian females had been taken as slaves by supposedly Papuan Y-DNA-bearing, indigenous males, then the resulting population would not be speaking Austronesian languages (much less *solely* Austronesian languages).

The same applies to your second hypothetical scenario, according to which a small group of Papuan "warrior males" would have crossed over from New Guinea to the Admiralty Islands subsequent to the establishment of a small original population of Austronesian colonists, and eliminated most of the original Austronesian male colonists from the Admiralties. In this case, too, as in the case of Latin America, the resulting population should be speaking the conquerors' Papuan language.

Why do you try to rationalize the researchers' obviously incorrect conclusions?

I much prefer terryt's hypothesis: an already "mixed" Austronesian-speaking population (with the origin of at least one of the constituents, i.e. the one bearing haplogroup C2 Y-DNA, being indeterminable) colonized the Admiralty Islands, and this population and its Austronesian language(s) have survived from the time of colonization down to the present day.

Dragon Horse said...


you have a point about the language spoken...yes, i think terry is on to something as well.

terryt said...

I've made quite a study of Polynesian origins since I studied their music when I did ethnomusicology more than 20 years ago. And I live in New Zealand. Here are links to a couple of essays I've had posted at remotecentral. Hope you find them interesting. I had some discussion with a linguist over my language diagram but I think we agreed it basically stands.



They're rather long but they both fit in with more recent work that claims a connection between Ket and Na-Dene.

terryt said...

By the way. The original intent of the essays was to explain how we should interpret more ancient evidence for our evolution. Replacement of mtDNA and Y-chromosome haplogroups has been happening since way before modern humans apparently replaced Neanderthals.

South Central Haplo said...

I am very much curious about this subject . But the terms being used are too generic. They confuse me some times.

Now a days the word Asian is being hijacked for Chinese/East Asians by popular media.

Austronesians: This term is being used for people from India to Australia.
Polynesia, Melanesia, Austranesia.

Again Indonesia with lot of diversity and common border is not part of this lot of times

Then it seems Admiralites are Austranesian speaking Melanesians.

What exactly this Austranesian speaking means?.

Terry and Dragon: Please feel free to enlighten me.

Whether in Indonesia or Melanesia or Polynesia.

Language replacement and genetic replacement are different. Some time they are proportionate and some time not necessarily.

Lets see genetic anthropology:

The original inhabitants in all these places are dark. I strongly believe. say 20000 -5000 years ago.

Their genetic markup has any O*?. I don't think so.
Where does these K* came up with out no trace of continuity. More research need to be done.
But they do have Y haplo C*, Mt haplo M, R etc.

Then there is this positive selection. For any society the norm for the last 1000 or more years is to bring in fair skinned bride . not the vice versa.
As king goes the common man.

Whether it is Pakistan or India or Polynesia the dark bride will be discriminated when there is this other availability.

The authors are trying to put together some genetic groupings which were seen else where it seems.

terryt said...

South central haplo. I agree we should differentiate between East Asia, South Asia, Southeast Asia and Melanesia/Australia. The regions have completely different-looking people.

As for the rest of you queries, we'll start with Polynesians and work back. Polynesians originally lived in the triangle between Hawai'i, Easter Island and New Zealand. Some had spread back west but we'll ignore them. Polynesian refers to both the language and the people, and Polynesia is the region.

The Polynesian language is classified as Austronesian, which is totally a language. It's technically incorrect to use the term for a people unless you use the term Austronesian-speaking people. Austronesian languages are spread from Polynesia in the east to Madagascar in the west. The languages spoken through much of SE Asia, including Indonesia and Malaysia, are also Austronesian. just some of the languages spoken in Melanesia are Austronesian, mostly those in coastal regions and in Southern Melanesia.

The words Melanesia and Micronesia are used to define regions, although people in those regions are often called Melanesian or Micronesian. There is no such thing as a Melanesian or Micronesian language group. Languages spoken in the regions belong to several groups. Melanesia refers to the region from Fiji in the east to New Guinea in the west. Micronesia refers to the many small islands north of this region.

You are therefore totally correct when you say, "Then it seems Admiralites are Austranesian speaking Melanesians". You are also correct, "Language replacement and genetic replacement are different" as is shown in Melanesia.

South Central Haplo said...

Thanks Terry.

More curious about ebizur comments. I agree his point about C2 concentration.

Either authors or Ebizur can not make that the basis for Language or Genes.

The article talks about disproportionate mt DNA and like to hear his comments on that.

bluejay said...

How do we download the Y chromosome data? sorry for asking such a dumb question, but have been trying for a week now, and trying to trace back the kayser paper, it goes back to 2000, and thats it, there is no accession numbers, nothing. Can anybody help me with this?