NEW YORK (GenomeWeb News) – In the early, online edition of the American Journal of Human Genetics yesterday, a research team describes how it took advantage of ancient interbreeding between humans and archaic hominin forms to reconstruct historical human migrations into Asia and Southeast Asia.
The work, which built on prior studies uncovering evidence of admixture between modern humans and archaic hominins such as Neandertals and Denisovans, involved scouring present-day human genomes for Denisova sequences. By determining the presence and proportion of Denisovan DNA in individuals from dozens of Southeast Asian and Oceanic populations, researchers were able to parse out previously inaccessible details about human population and migration history.
The patterns they detected — including evidence that Denisovan ancestry is restricted to populations from eastern Southeast Asia and Oceania — points to a two-stage peopling of the region, consistent with conclusions reached by authors of an Aboriginal Australian genome study published in yesterday's issue of Science.
"I think it does provide pretty strong support for multiple dispersals," senior author Mark Stoneking, an evolutionary genetics researcher at the Max Planck Institute for Evolutionary Anthropology, told GenomeWeb Daily News.
When an international research team sequenced the Denisova genome from a 30,000 to 50,000-year-old finger bone fragment from a cave in Siberia, they found that the ancient hominin shared some sequences with individuals from Melanesian populations. But the timing and extent of this admixture was poorly understood due to scant sampling of populations from Oceania and Southeast Asia at the time.
Authors of the current study had access to many samples from populations in this region, Stoneking explained, and decided to look for evidence of gene flow from Denisovans in more detail.
"What we were mostly interested in doing when we started the study was just to see what populations in the area might show signatures of Denisova gene flow and then what, if anything, that might tell us about human dispersals," he said.
The researchers genotyped 260 individuals from 33 populations including East Asian, Southeast Asian, New Guinean, Fijian, Polynesian, Australian, and Indian with the Affymetrix SNP 6.0 arrays. They also used the Illumina GAIIx to sequence the genomes of two individuals: a highlander from Papua New Guinea and an individual from the Mamanwa Negrito population in the Philippines.
The team did not detect any evidence of Denisovan admixture in populations from Western Indonesia or mainland East Asia, but did find Denisovan sequences in an individual from New Guinea and in samples from the Mamanwa group from the Philippines, Aboriginal Australians, and populations from Eastern Indonesia, Fiji, and Polynesia.
Because other studies have hinted that some of these populations contain gene flow from New Guinea, Stoneking explained, researchers decided to look at the proportion of New Guinean ancestry to see whether it coincided with the amount of Denisovan sequence turning up in populations.
"The prediction would be that if New Guinea ancestry explains Denisovan ancestry, then the amount of New Guinea ancestry should be correlated with the amount of Denisovan ancestry," he said.
Indeed, the results were consistent with a model in which Denisovan sequences entered Eastern Indonesia, Fiji, and Polynesia populations as a consequence of gene flow from Papua New Guinea.
In contrast, though, the investigators found that Denisovan sequences in the Mamanwa population were independent of New Guinean ancestry. Genetic patterns in that population did not point to Denisovan gene flow via New Guinean or Australian Aboriginal populations.
Together, the results suggest that there were two migrations into the regions, with Denisovans interbreeding with the ancestors of the modern humans who ended up in parts of the Philippines, Australia, and New Guinea.
"There was a single period of gene flow from Denisova into a population that was ancestral to the Philippine Negrito, the Australians, and the New Guineans," Stoneking said. "After that Denisovan gene flow, these populations diverged. First the Philippine Negrito groups diverged and then the Australians and New Guineans diverged from a common ancestor."
From the data available so far, it is difficult to say when the admixture between Denisovans and humans occurred. But since it seems to have happened in a common ancestor to the Australian, New Guinean, and Negrito populations, Stoneking explained, it is thought to have preceded the split between Australian and New Guinean populations an estimated 50,000 years ago.
"We are trying to date it directly using genomic data, but that so far is a pretty tough problem," he said. "We're still working on that."
Determining where the inter-breeding with Denisovans occurred is no small task either, though there are clues that it may have been in eastern Southeast Asia: researchers have not found evidence of Denisovan admixture in two Malaysian and Andaman Island populations that are thought to have descended from the same ancestral Negrito group as the Mamanwa, suggesting human-Denisovan breeding might have happened after the first modern humans arrived in Asia and started splitting off into different areas.
If so, Stoneking said, Denisovan distribution may have been unexpectedly far-reaching. "If we do accept that the Denisova gene flow occurred in eastern Southeast Asia, that would suggest that they were very widespread, both geographically and ecologically," he said.
The team is continuing to try to narrow in on the timing of Denisovan gene flow into modern humans. They are also interested in looking at whether Denisova genes have any functional roles in populations where they are still found today.
Such studies should become somewhat easier as higher quality Neandertal and Denisova genomes become available, Stoneking said, noting that more genomic data from archaic hominins could also help pinpoint more subtle evidence of archaic admixture in human genomes.