NEW YORK – Researchers have teased out evidence suggesting the ancestors of modern-day sub-Saharan Africans interbred with an archaic human group. At the same time, though, they found no evidence of non-Neanderthal or non-Denisovan archaic admixture among Andaman Islanders, which had been previously suggested.
As anatomically modern humans migrated from Africa, they came into contact and interbred with archaic human groups. Non-Africans are estimated to have inherited about 2 percent of their DNA from Neanderthals, while Melanesian and aboriginal Australians have inherited about 4 percent to 5 percent of their DNA from Denisovans, and yet other populations, like the Andaman Islanders, were thought to have interbred with as-yet-unknown hominins. But as archaic humans also lived in sub-Saharan Africa around the same time and in the same places as a modern humans, there may have been opportunities for admixture inside Africa, as well.
To study this, Jeffrey Wall, a researcher at the University of California, San Francisco and his colleagues used a dataset of genomes from more than 1,600 individuals from across the globe, recently reported by the GenomeAsia 100K Consortium, to search for signs of archaic admixture. As they reported today in the American Journal of Human Genetics, the researchers found evidence of such admixture among sub-Saharan African individuals based on a linkage disequilibrium approach. They did not find, though, evidence for interbreeding with archaic humans among Andaman Islanders or people from the island of Flores.
"We note that the extent of admixture between modern and archaic humans, while clearly of historical and evolutionary interest, also has direct implications for human disease genetics," Well and his colleagues wrote in their paper.
The researchers searched for signs of admixture with archaic human groups without relying on genome sequence data from those groups. Instead, they used a linkage disequilibrium approach and searched for long, diverged haplotypes within the 1,667 genomes from the GenomeAsia 100K Project (GAsP) pilot dataset, which includes not only individuals from Asia but also from Africa, Europe, and the Middle East.
The researchers considered the long, diverged haplotypes they found as candidate regions for archaic human introgression and dubbed them "putative ghost haplotypes," or PGHs.
After validating their approach by searching for known Denisovan-derived tracts within non-African samples from GAsP, the researchers then applied it to their full GAsP cohort. Through this, they identified 2,319 autosomal PGHs.
The average number of PGHs, they noted, was much higher — between five and 15 times — among sub-Saharan Africans than non-Africans. It was also higher than the researchers had expected based on simulated datasets.
This suggested that a major archaic admixture event may have taken place in sub-Saharan Africa. Based on differences in PGHs among genomes from Khoesans, Central African Pygmies, and West, East, and North Africans, they further said that this admixture event likely took place in the ancestors of modern-day Khoesans before spreading to other modern human populations.
The researchers cautioned, though, that their approach cannot fully disentangle models of long-term isolation and population structure from their proposed archaic admixture scenario, and said that additional sequencing and analysis of genomes from Central and Southern African groups will be needed.
Meanwhile, they found few signs of non-Neanderthal, non-Denisovan archaic admixture among non-Africans, including South Asians and Melanesians. A previous study had reported a signal of unknown archaic admixture among Andaman Islanders, but the UCSF-led team could not detect any, despite deep sampling in the region. Additionally, they found no signs of additional archaic admixture among individuals from Flores, where the ancient hominin Homo floresiensis was uncovered.
The researchers also found that the PGHs they uncovered were generally found away from coding regions and appeared to be under purifying selection. This, they noted, is in line with recent findings suggesting that introgressed Neanderthal regions are harmful in modern humans and may contribute to a number of common diseases.