Analysis of DNA from ancient humans has revealed its power decisively in the last few years, and especially at the beginning of 2014 with publication of the sixth full genome of an individual who was not an anatomically modern human (Prüfer, K. and 44 others 2014. The complete genome sequence of a Neanderthal from the Altai Mountains. Nature, v. 505, p. 43-49). The newly sequenced material came from a toe bone found in the Denisova Cave in the Altai Mountains of southern Siberia; the same location made famous in 2010 by genetic evidence for unknown late hominins, the Denisovans . The bone occurred in the same layer of cave sediment, dated at 50.3 ka, which yielded the Denisovan finger bone, but from a lower sublayer. So there is no firm evidence that both groups cohabited the cave.
The genome reveals that the individual was female and related to the three Neanderthals from Croatia and another infant Neanderthal from the Caucasus, also analysed previously by Svante Pääbo’s team at the Max Planck Institute for Evolutionary Anthropology in Leipzig, Germany (Note that the toe-bone team also includes co-workers from US, Chinese, Austrian, French and Russian institutions). The closest statistical link is to the Caucasian infant Neanderthal’s DNA. Interestingly, it proved possible to demonstrate that the Siberian Neanderthal woman was from a population that was clearly inbred, her parents having been related at the level of half siblings. Her mtDNA shows that she shared a common ancestor with all 6 Neanderthals from whom mtDNA has been analysed.
Comparing genomes from the single Denisovan, the 5 Neanderthals and living humans from sub-Saharan Africans gives an estimated 550 to 765 ka time of divergence of a population leading to anatomically modern humans from the progenitors of Neanderthals and the Denisovan. The Neanderthal-Denisovan split was roughly 380 ka ago. It was already known that non-African living humans contain genetic evidence for past interbreeding with Neanderthals and that some people in Asia, Australia, Melanesia and the Philippines had acquired genes from Denisovans. More refined comparisons now show Oceanians to have 3 to 6% Denisovan make-up with Asians in general sharing 0.2%. Neanderthal to modern non-African gene flow is now estimated at between 1.5 and 2.1%, with Asians and Native Americans being at the high end. Neanderthals and Denisovans also interbred, but only at the level of about 0.5% inheritance. However, that genetic sharing involved DNA regions known to confer aspects of immunity and sperm function, that also made their way into living non-African humans.
Since the common ancestor of Neanderthals and Denisovans left Africa long before modern humans appeared on the scene it would be expected that living Africans’ genomes would show the same level of similarity with both the now extinct groups, if all three originally shared a common ancestor. A surprising outcome from comparison of Neanderthal and Denisovan genomes with those of living sub-Saharan Africans is that there is a significant bias towards Neanderthal rather than Denisovan comparability. There are three possibilities for this bias. After the Neanderthal-Denisovan split the former group may have continued to interbreed with the group leading to modern Africans (and indeed to modern non-Africans): that would require Neanderthal genetics to have originated in Africa before they migrated to Eurasia. Secondly, the gene flow could have been from the ancestors of modern humans to Neanderthal progenitors, making descendant Neanderthals more like modern humans. Prüfer et al. suggest that the evidence is less supportive of both and weighs towards a third possibility; that the Denisovans interbred with an unknown contemporary hominin, whose genetic make-up was yet more different from that of all three known groups of the late Pleistocene and therefore their common ancestor . This may have been Homo antecessor or possibly H. erectus who survived until as late as 20 ka in SE Asia.
As other commentators on the paper (Birney, E. & Pritchard J.K. 20113. Four makes a party. Nature, v. 505, p. 32-34) have observed, ‘…Eurasia during the late Pleistocene was an interesting place to be a hominin, with individuals of at least four quite diverged groups living, meeting and occasionally having sex.’ All this arises quite convincingly from the genetics of only 7 ancient individuals, to show that it may no longer be appropriate to consider human evolution as a tree or a bush linking permanently separated species. Either it is the history of a single, polymorphic species – remains of 1.7 Ma old Homo georgicus show clear evidence of such polymorphism – or a better metaphor for human development is an interwoven basket or twine. Rumour has it that attempts are being made to sequence an H. antecessor dated at 900 ka from Gran Dolina Cave in the Atapuerca Mountains in Northern Spain: as they say, ‘Watch this space’!