search this blog

Saturday, October 31, 2015

Aboriginal Australian Y-chromosomes


Based on a survey of 657 self-declared Aboriginal individuals:

C-M130*(xM8,M38,M217,M347) 1%
C-M347 19%
K-M526*(xM147,P308,P79,P261,P256,M231,M175,M45,P202) 12%
S-P308 12%
M-M186 0.9%
Non-indigenous 56%

Unlike an earlier paper using genome-wide DNA, this study found no signals of gene flow from India to Australia during the Holocene.

Source...

Nagle, N., Ballantyne, K. N., van Oven, M., Tyler-Smith, C., Xue, Y., Taylor, D., Wilcox, S., Wilcox, L., Turkalov, R., van Oorschot, R. A.H., McAllister, P., Williams, L., Kayser, M., Mitchell, R. J. and The Genographic Consortium (2015), Antiquity and diversity of aboriginal Australian Y-chromosomes. Am. J. Phys. Anthropol.. doi: 10.1002/ajpa.22886

7 comments:

andrew said...

The Y-DNA M and S have historically been seen mostly in Papuans (and neither are basal at all). C* is usually seen, as here, with K and these are almost never found in the company of D. C-M437 is a reasonably basal derivative of C*.

The hard question is whether the overlap with Papuan Y-DNA is a result of drawn from overlapping gene pools in the respective founding populations, or a later substantial introgression of Papuan Y-DNA into Australian aborigines.

It would also be interesting to know how much of the non-indigenous DNA is Asia v. European.

capra internetensis said...

@andrew

This is S-P308, which Karafet et al (2015) observed only in Australia, a sister to the main branch S-M230, which is Melanesian. Where exactly P308 falls in the phylogeny is unclear but according to Karmin et al the two oldest splits in S are only 2 + 1 SNPs (a few centuries) after the TMRCA of MS itself.

So to answer your question, I'd say that most of the overlap is indeed from the founding populations and not later gene flow.

Gaspar said...

Interesting it showed that K-M526 most likely are all K-P60 as per Karafet 2014 data

http://www.nature.com/ejhg/journal/v23/n3/fig_tab/ejhg2014106t1.html

terryt said...

"Interesting it showed that K-M526 most likely are all K-P60 as per Karafet 2014 data"

Interesting that, once again, K2b1a1-P60, shared with New Guinea, is shown to be not as common as C. The other branches of K2b1a-P405 are all either Eastern Indonesian or Papuan.

"The hard question is whether the overlap with Papuan Y-DNA is a result of drawn from overlapping gene pools in the respective founding populations, or a later substantial introgression of Papuan Y-DNA into Australian aborigines".

I very much suspect the latter, although I haven't had time yet to read the paper and see where S and M were found. It would make sense if they were from Cape York or some other northern region.

"This is S-P308, which Karafet et al (2015) observed only in Australia, a sister to the main branch S-M230, which is Melanesian".

Doesn't mean it was part of the founding population.

"C-M437 is a reasonably basal derivative of C*".

C1b2b-M347 (I presume you meant that) is found only in Australia as far as I'm aware. C1b2a is found along the northern coast of New Guinea and out into the Pacific. It almost certainly first developed in the Lesser Sunda Islands.

terryt said...

A bit more:

"Our data support the hypothesis of more than one route (via New Guinea) for males entering Sahul some 50,000 years ago"

It is very difficult to make a case for Australian Y-DNA C having entered via New Guinea. New Guinea is remarkably free of Y-DNA C other than C1b2a, the oldest dates for which are from Southern Wallacea. Its presence in New Guinea may even be as recent as the Austronesian expansion. Australian Y-DNA C looks to have entered direct from Timor. I agree the others look very much to be derived from New Guinea haplotypes.

"Haplogroups C-M347, K-M526*, and S-P308 are Aboriginal Australian-specific".

I agree C-M347 is 'Aboriginal Australian-specific', as may be S-P308. But unless K-M526 is a branch within K2b1a1-P60 it is not.

"Where exactly P308 falls in the phylogeny is unclear but according to Karmin et al the two oldest splits in S are only 2 + 1 SNPs (a few centuries) after the TMRCA of MS itself".

I see no mention in ISOGG but I notice Wikipedia has this to say concerning S in general:

"One study has reported finding haplogroup S-M230 in: 52% (16/31) of a sample from the Papua New Guinea Highlands; 21% (7/34) of a sample from the Moluccas; 16% (5/31) of a sample from the Papua New Guinea coast; 12.5% (2/16) of a sample of Tolai from New Britain; 10% (3/31) of a sample from Nusa Tenggara, and; 2% (2/89) of a sample from the West New Guinea lowlands/coast.(Kayser 2003 Cox 2006)"

And so it is present in parts of Wallacea adjacent to both New Guinea and Australia. Therefore it too may have entered Australia direct from Timor.

capra internetensis said...

@Gaspar

If you look at the supplementary information for the paper it breaks down the K2 further and you can see that about half of it is indeed what Karafet called K2b1a1-P60 and what this paper calls S4-P308 - P308 and P60 are equivalent SNPs so it is the same thing. However, the remainder of the Australian K* is classified as K2*-M256(xM147, P308, P79, P261, P256, M231, M175, M45, P202) which rules out S-M230 (S1 in the new nomenclature), M-P256, N, O, QR, and several other branches of basal K2, but not all of the known branches.

So it is not all P60. It's just unclassified, we can't say whether it is one clade or several, whether it is related to NO or MS or P or none of them, where else it might be found, etc.

@Terry

K-P60/S4-P308 is not found in New Guinea, nor S/S1-M230 in Australia, as far as I know. The various branches of pre-S/S-P405 are pretty widespread though predominantly Melanesian. There is not enough evidence to say whether P60/P308 was a very basal split or a more recent one or whether it came to Australia via Indonesia or New Guinea - both seem equally parsimonious.

terryt said...

"P308 and P60 are equivalent SNPs so it is the same thing".

And:

"K-P60/S4-P308 is not found in New Guinea"

Yes, I've checked the Karafet paper and it says:

"K-P304 and K-P308 are observed only in Australia (and are equivalent to K-P60)".

"There is not enough evidence to say whether P60/P308 was a very basal split or a more recent one or whether it came to Australia via Indonesia or New Guinea - both seem equally parsimonious".

Both the Karafet paper and ISOGG have K-P60/P304/P308 being one of four branches within K2b1a. The other three being found in Melanesia/New Guinea. Consequently I think we can assume that the Australian K2b1a1 arrived at a time when K2b1a was expanding from New Guinea, not from eastern Indonesia. On the other hand K2b1a itself is again one haplotype within four. The other three being in fact eastern Indonesian and Filipino. That suggests the New Guinea/Melanesia K1b1a (its ancestor anyway) arrived as part of an expansion across Wallace's line.

Overall the Y-DNA K2's phylogeny indicates a series of separate expansion. First is an original expansion south along the SE Asian island chain at a time when the islands were connected to each other. Then a branch of K2b crossed Wallace's Line and formed K2b1 (I'll ignore K2b2 at this stage). K2b1 reached Borneo, the Philippines and New Guinea, with some remaining in Nusa Tenggara. In the separate regions it formed K2b1b, K2b1c, K2b1a and M respectively. Then K2b1a expanded from New Guinea into Melanesia and Australia, and even back into eastern Indonesia, whereupon it formed the various K2b1a haplotypes.