37,000-year-old Russian skeleton has Neanderthal DNA that’s gone missing

Another week, another ancient human genome. We just recently covered the oldest modern human genome yet described. Now, another paper takes a look at the DNA from a different modern human genome and comes to similar conclusions: interbreeding with Neanderthals was already deep in the past as of 37,000 years ago. But researchers were able to find stretches of the Neanderthal genome that are no longer present in any modern human populations that we've sampled.The skeleton in this case comes from the European area of Russia; it was found at a site called Kostenki-Borshchevo north of the Black Sea. The team behind the new paper (which does not include Svante Pääbo, who has pioneered ancient genomics) was only able to get a rough draft of the individual's genome, on average sequencing every base 2.4 times. Thus, the sequence is likely to include a large number of errors and gaps. These make the conclusions a bit more tenuous than previous work but shouldn't bias them in any particular direction.

One thing the results make clear is that humanity's migration out of Africa was complicated. K-14, as the skeleton is called, shares very few of the DNA differences that are associated with East Asian populations, as has been the case with the Siberian modern human skeletons we've looked at. All of which suggests that East Asians and Eurasians split off early and may even have engaged in separate migrations out of Africa or the Middle East. K-14 also lacks common variants found in Native Americans, leaving a single Siberian skeleton as the only one that has an affinity to them.

Despite its location, K-14 also lacks a strong genetic connection to modern Europeans, instead having a general affinity for other early Eurasian populations. In fact, the authors conclude, it may not even make sense to look for specific affinities. "Instead of inferring a few discrete migration events from Asia into Europe," the authors write, "we now see evidence that humans in Western Eurasia formed a large meta-population with gene flow in multiple directions occurring repeatedly and perhaps continuously."

In other words, don't expect to find a couple of populations that were the European ancestors; instead, there was a large pool of Eurasian populations that regularly intermingled.

Speaking of intermingling, we have the Neanderthals. Just as with the recent Siberian results, the absolute percentage of Neanderthal DNA was similar in K-14 and current human populations. But the length of the average stretch of Neanderthal DNA was longer, suggesting that there had been less time for recombination to scramble these sequences. The authors used this to estimate the time when interbreeding took place, and they come up 54,000 years ago—very similar to the 60,000-year figure estimated using the ancient Siberian DNA.

The authors performed one other test involving Neanderthal DNA: identifying the areas where current human populations lack Neanderthal DNA and seeing if any samples from ancient skeletons have it there. Most individuals have nothing; about one percent of K-14's Neanderthal DNA comes from these regions, suggesting that, in the intervening 37,000 years, these stretches of Neanderthal DNA have either been selected against or simply lost by random chance.

The sequencing of ancient genomes is now clearly a competitive field. In fact, last week's paper on the Siberian skeleton came out while this paper was still in review, suggesting Science rushed to get it into print while it was still considered relevant. It's a reasonable fear; as similar results pile up, it's likely that each further advance won't be considered as newsworthy.But the cumulative weight of these and other results may ultimately be more important than most of the earlier finds. Some aspects, like the ancestry of current Europeans, aren't becoming any clearer with more samples, suggesting that the ancestry itself is confused. Others, like the Native American affinities found in the Altai skeleton's genome, might suggest there was a large overlap between Eurasians and Native Americans. Now, with more genomes, it now looks like this skeleton is a rare exception.

So in a few decades, when textbooks are written about humanity's journey out of Africa, the story will probably be built from the results that appeared long after the papers that made headlines.