This is a technical paper, so the devil is in the details; I'll post any technical comments in this space, after I read it carefully. But, I want to highlight three things:
- The date of admixture is consistent with a pattern of late Neandertals (including Vindija, the source of the Neandertal genome, and also a few in the Near East, such as Kebara, Amud, and Shanidar) becoming more similar to modern humans over time. This does not speak of occasional admixture -which wouldn't have affected all Neandertal remains- but of thorough inter-deme gene flow.
- The date of admixture is inconsistent with the coastal migration theory c. 60,000 years ago, because there is no evidence for Neandertal occupation of the coast (either in terms of lithics or anthropology)
- The authors assume that the admixture event corresponded to the spread of Upper Paleolithic carrying humans Out of Africa at the time it occurred. But, there is really no evidence that the Upper Paleolithic originated in Africa, and its earliest attestation I could find was in the Levant.
The period of the inferred Neandertal admixture coincides with a period of climatic deterioration in Arabia. It would be an awful time for humans to be migrating from Africa to Asia; this is why the coastal migration theory was proposed in the first place: to give migrating humans plentiful marine resources as they ventured further away from Africa. But, the evidence does not square: modern humans migrating along the coast would not have an opportunity to admix with Neandertals and they would not have the opportunity to influence Neandertals at sites in Israel and Iran to appear more modern at this time.
- there is every reason to think that modern humans thrived during wet periods in North Africa and the Levant (~135-115 ka and ~105-75 ka).
- the post-70ka climatic deterioration is a really bad time for people to be venturing into Asia out of Africa; the coastal migration theory is off, because it doesn't explain Neandertal admixture or the transformation of Asian Neandertals well away from the coast
- the post-70ka climatic deterioration is precisely the type of ecological crisis that would spur a colony of modern humans, who had been bottled up in their green Arabian peninsula before that time, to venture north, away from their deteriorating homeland, bringing them in closer contact with the Neandertals, and beginning the grand merge of the two populations.
They were loved out of existence, absorbed into the population of migrating Out-of-Arabians fleeing an ecological catastrophe affecting their land, c.70ka.
The signs of that episode of inter-palaeo-deme fusion exist in the morphology of the late Neandertals: not diverging away from modern humans, but already noticeably modern-leaning themselves. And, they exist in the genomes of modern Eurasians who preserve a little more (but how many?) of their genes than Africans do.
And, it was perhaps here that the quantum leap in human behavior began; perhaps a case of punctuated equilibrium, as the Middle Paleolithic Arabians were hit hard by ecological disaster. Perhaps a case of hybrid vigor as MP Arabians and MP Neandertals mixed their genes and hit a kind of evolutionary jackpot. Or, perhaps, too, something as simple as a couple of hunters comparing toolkits, either in peace or in battle, encountering people different than themselves, and finding, through synergy or competition, the beginnings of a New Idea.
An important technical caveat about the paper is that they have chosen a 10% minor allele frequency cutoff for SNPs. It is well known, however, that MAF is correlated with time, with low-frequency polymorphisms generally being more recent than high-frequency ones. A new variant -excepting the case of balancing or frequency-dependent selection- tends to either become extinct or fixed over long periods of time. Conversely, recent variants tend to occur at low frequencies.
It would certainly be interesting to see what admixture estimates look like, if we only take low MAF SNPs into consideration. Note, that the published admixture estimates of a few percent were genome-wide averages without MAF-filtration.
This is important, because these age estimates might be influenced by older gene flow (SNPs with more than 10% minor allele frequency) and also by structure in Africa. As I have explained before, Eurasian-shift towards Neandertals, relative to Africans, can occur if Africans have admixture with a Homo lineage before the last common ancestor of modern humans and Neandertals. In those cases, the minor allele may be "archaic African", but such sites would also contribute to a perceived Neandertal-shift of Eurasians relative to Africans.
Finally, LD-based age estimates can only take you so far, because LD decays over time. Older episodes of gene flow cannot be captured by such methods. So, I see no reason to doubt the idea that Neandertals never really evolved as a separate lineage from modern humans, and rather represented a palaeo-deme that was not genetically isolated from African humans. The results of this paper do suggest that there was a major recent episode of gene flow that eventually blended the two demes (Arabian + Neandertal) just prior to the onset of the Upper Paleolithic, but we cannot really exclude older episodes of gene flow.
UPDATE II (Aug 14): Another paper on the topic which casts doubt on substantial modern-Neandertal admixture by Eriksson and Manica.
The date of interbreeding between Neandertals and modern humans
Sriram Sankararaman, Nick Patterson, Heng Li, Svante Paabo, David Reich
Comparisons of DNA sequences between Neandertals and present-day humans have shown that Neandertals share more genetic variants with non-Africans than with Africans. This could be due to interbreeding between Neandertals and modern humans when the two groups met subsequent to the emergence of modern humans outside Africa. However, it could also be due to population structure that antedates the origin of Neandertal ancestors in Africa. We measure the extent of linkage disequilibrium (LD) in the genomes of present-day Europeans and find that the last gene flow from Neandertals (or their relatives) into Europeans likely occurred 37,000-86,000 years before the present (BP), and most likely 47,000-65,000 years ago. This supports the recent interbreeding hypothesis, and suggests that interbreeding may have occurred when modern humans carrying Upper Paleolithic technologies encountered Neandertals as they expanded out of Africa.