Thursday, 16 August 2012

Neandertal STAT2 haplotype in Eurasians

Two recent papers have argued that African population structure or late Middle Paleolithic/Upper Paleolithic Neandertal admixture have contributed to the finding that Non-Africans appear to be a few percent more similar to Neandertals than Africans are across the genome. I would add that modern human admixture in the Vindija individual remains a distinct possibility.

What percentage of the ~3% Eurasian excess can be accounted by each of these three processes? The jury is out, and we won't find out until someone decides to tackle the problem comprehensively and/or new ancient DNA samples become available to inform the discussion. African population structure cannot be discounted, and intriguing new evidence may appear thanks to ancient DNA analysis.

But, there is a different approach to detecting Neandertal admixture that zeroes in on specific genomic locations and dissects them in great detail. This single-region approach provides evidence for admixture, without necessarily arguing about how extensive it was.

The single-region dissection was previously used in the Hammer lab to identify the first very convincing evidence for archaic admixture in Africans and Melanesians. In a new paper, Mendez et al. identify a small region in chromosome 12 that shows evidence for archaic introgression from Neandertals, or a species closely related to them.

But, it is worthwhile to begin with a list of other Neandertal introgression candidates from the literature:

Thus far, only a handful of loci have been hypothesized to have entered the human gene pool through archaic admixture and positive selection, including MAPT (MIM 157140),5 MCPH1 (MIM 607117),3 and particular alleles at the HLA locus (MIM 142800, 142830, 142840).6 However, analysis of the Neanderthal genome failed to provide evidence of introgressive alleles at the former two loci.1 Because of its role in fighting pathogens, HLA presents an instance where it is relatively easy to conceive of an a priori reason that acquisition of an archaic Eurasian HLA allele would benefit human ancestors, especially as they expanded into new habitats.7 However, the fact that HLA haplotypes are known to exhibit transspecific polymorphism and show evidence of strong balancing selection 8,9 increases the probability that similarities between modern and archaic haplotypes are due to ancestral shared polymorphism (i.e., as opposed to archaic admixture). In addition, the SNPs tagging the main HLA haplotype that was said to have introgressed were not observed in the Denisova or Neanderthal draft genomes. 
So, what lines of evidence support the notion that the new STAT2 haplotype is the "real deal"?
First, N matches the Neanderthal sequence at all 18 sites that fall within the resequenced 8.6 kb STAT2 region and have Neanderthal sequence coverage (Table 1). Second, N lineages are broadly distributed at relatively low frequencies in Eurasian populations (Figure 3) and are not observed in sub-Saharan African populations (Table S6). Third, the N haplotype extends for ~130 kb in West Eurasians and up to ~260 kb in some East Asians and Melanesians, producing much stronger LD than that observed in sub-Saharan Africans.


Given that the N lineage and the reference sequence diverged ~600 kya, these results suggest that population structure has influenced the recent evolution of this locus. Balancing selection alone is not expected to maintain this extent of LD and consequently is not sufficient to explain these patterns. Moreover, although a strong bottleneck could generate extended LD similar to the levels we observe near STAT2 in non-Africans, it would not explain why the N lineage went extinct in Africa (i.e., why the SNPs associated with the N lineage in non- Africans were not observed in sub-Saharan Africans that are part of our WGS or public SNP panels).


We point out that although a recent common ancestry between a human lineage and Neanderthal sequences might indicate gene flow between Neanderthals and modern humans, this information alone does not inform us about the direction of gene flow. With the additional evidence of the observed extent of LD in modern human sequences, it is possible to infer that the N lineage introgressed into modern humans (either from Neanderthals or another archaic source that contributed to both Neanderthals and AMH).
Actually, the N haplotype is observed in North Africa, but this might be due to relatively recent back-migration. One might also argue that a recent bottleneck in a Eurasian population generated the high degree of LD, and the N haplotype was lost in a back-to-Africa migration, or North-to-Sub-Saharan Africa migration. But, that would not seem to explain how the deeply divergent lineage persisted in the North African population of proto-modern humans for such a long time; the evidence for recent common ancestry of N with the Neandertal haplotype would argue against incomplete lineage sorting (=inheritance of related forms of the haplotype from before the modern-Neandertal divergence).

All in all, this probably represents the best evidence for Neandertal-to-modern introgression to date. As full genomes of different human groups become available, it will be possible to automate this analysis and pick off other such strong signals. This may not indicate the level of admixture, but it might provide strong evidence against the idea of reproductive isolation between modern humans and Neandertals.

It is also noteworthy that this is barely consistent with the coastal migration theory with respect to the origin of Australo-Melanesians, because humans trekking along the coast would not have the opportunity to admix with Neandertals who are completely unattested there in either their physical, or archaeological (Mousterian) form.

But, it is consistent with my Out-of-Arabia theory. Australo-Melanesian Y chromosomes belong to the CF clade of the phylogeny. I have speculated that the post-70ka climate crisis in Arabia spurred some human groups to escape north (CF), and others to remain south (DE). The latter eventually gave rise to the major African lineage, heading west (E), as well as a relic Asian lineage heading east (D) that was later inundated by the descendants of CF. If Australo-Melanesians are descended from the CF folk who went north out of Arabia, then they too would have had the opportunity to admix with Neandertals in the Near East.

The American Journal of Human Genetics, Volume 91, Issue 2, 265-274, 10 August 2012

A Haplotype at STAT2 Introgressed from Neanderthals and Serves as a Candidate of Positive Selection in Papua New Guinea

Fernando L. Mendez, Joseph C. Watkins and Michael F. Hammer

Signals of archaic admixture have been identified through comparisons of the draft Neanderthal and Denisova genomes with those of living humans. Studies of individual loci contributing to these genome-wide average signals are required for characterization of the introgression process and investigation of whether archaic variants conferred an adaptive advantage to the ancestors of contemporary human populations. However, no definitive case of adaptive introgression has yet been described. Here we provide a DNA sequence analysis of the innate immune gene STAT2 and show that a haplotype carried by many Eurasians (but not sub-Saharan Africans) has a sequence that closely matches that of the Neanderthal STAT2. This haplotype, referred to as N, was discovered through a resequencing survey of the entire coding region of STAT2 in a global sample of 90 individuals. Analyses of publicly available complete genome sequence data show that haplotype N shares a recent common ancestor with the Neanderthal sequence (∼80 thousand years ago) and is found throughout Eurasia at an average frequency of ∼5%. Interestingly, N is found in Melanesian populations at ∼10-fold higher frequency (∼54%) than in Eurasian populations. A neutrality test that controls for demography rejects the hypothesis that a variant of N rose to high frequency in Melanesia by genetic drift alone. Although we are not able to pinpoint the precise target of positive selection, we identify nonsynonymous mutations in ERBB3, ESYT1, and STAT2—all of which are part of the same 250 kb introgressive haplotype—as good candidates.