When modern humans migrated out of Africa some 60,000 years ago, they found the Eurasian continent already inhabited by brawny, big-browed Neanderthals. We know that at least some encounters between the two kinds of human produced offspring, because the genomes of people living outside Africa today are composed of some 1 to 4 percent Neanderthal DNA.
Some parts of non-African genomes are totally devoid of Neanderthal DNA, but other regions abound with it, including those containing genes that affect our skin and hair. This hints that the Neanderthal gene versions conferred some benefit, and were kept during evolution.
"It seems quite compelling that as modern humans left Africa, met Neanderthals, and exchanged genes, we picked up adaptive variants in some genes that conferred an advantage in local climatic conditions," says Joshua Akey, who led the study in Science.
"The adaptive things from Neanderthals are very interesting because they are not obvious," says John Hawks of the University of Wisconsin-Madison, who was not involved in either study. Based on fossil bones alone, anthropologists would never have predicted that Neanderthals contributed to the keratin filaments and immune systems of modern people.
The fact that Neanderthal DNA is totally absent from other stretches of the modern non-African genome suggests that their versions of the genes in these regions would have caused problems in modern humans, and were weeded out by natural selection.
In the Nature study, Sriram Sankararaman and David Reich of Harvard Medical School used the previously sequenced Neanderthal genome to screen 1,004 modern genomes for sequences with distinctive Neanderthal features.
For example, if a fragment of DNA is shared by Neanderthals and non-Africans, but not Africans or other primates, it is likely to be a Neanderthal heirloom. Also, Neanderthal sequences are typically inherited in large batches, since they were imported into the modern human genome relatively recently and have not had time to break apart.
In the Science study, Akey and Benjamin Vernot, both of the University of Washington in Seattle, used similar statistical features to search for Neanderthal DNA in the genomes of 665 living people—but they initially did so without the Neanderthal genome as a reference. They still managed to identify fragments that collectively amount to 20 percent of the full Neanderthal genome.
Neanderthal Influence on Skin, Hair, Common Diseases
Despite their different approaches, both teams converged on similar results. They both found that genes involved in making keratin—the protein found in our skin, hair, and nails—are especially rich in Neanderthal DNA.
For example, the Neanderthal version of the skin gene POU2F3 is found in around 66 percent of East Asians, while the Neanderthal version of BNC2, which affects skin color, among other traits, is found in 70 percent of Europeans.
The Neanderthal version of these genes may have helped our ancestors thrive in parts of the world that they were not familiar with but that Neanderthals had already adapted to. "Neanderthals had been in these environments for hundreds or thousands of years," says Sankararaman. "As modern human ancestors moved into these areas, one way to quickly adapt would be to get genes from the Neanderthals."
"Unfortunately, skin and hair do so many things that it's hard to speculate on what specifically that adaptive trait was," says Akey.
Sankararaman also found Neanderthal variants in genes that affect the risk of several diseases, including lupus, biliary cirrhosis, Crohn's disease, and type 2 diabetes. The significance of these sequences is "even less clear."
Both teams found that non-African genomes have large continuous "deserts" that are totally devoid of Neanderthal DNA. These regions include genes such as FOXP2, which is involved in motor coordination and could play an important role in human language and speech.
The Neanderthal-poor deserts are especially big in the X chromosome, and include genes that are specifically activated in testes. This hints that some Neanderthal genes may have reduced the fertility of male modern humans and were eventually lost. However, Hawks cautions that this probably happened over hundreds of generations—it was very unlikely that the sons of Neanderthals and modern humans were obviously infertile.
DNA Hints at Other Mystery Humans
Both teams are now planning to apply their methods to other hominids like the Denisovans—an enigmatic group whose presence in Asia some 40,000 years ago is known just from DNA from a finger bone and some teeth found in a single cave in Russia.
And Akey's work shows that it may even be possible to partially reconstruct the genomes of unknown groups of ancient humans without any prehistoric DNA samples.
"That's one of the things that I'm most excited about," he says. "Paleogenomics is a difficult field because it often requires finding suitable fossils with well-preserved DNA. "Maybe we're not always beholden to bones. We can look at the genomes of present-day individuals."
It is becoming increasingly clear that the Pleistocene was awash with many different groups of early humans, hooking up with each other to various degrees. Recent studies, for instance, have found tantalizing hints of unknown groups from Asia and Africa that left genes in Denisovans and modern humans, respectively. Akey's method could give us our first glimpse at these mystery humans.
"If there is no fossil evidence and potentially never will be, this will be the only way of finding out about groups that were important in human history," he added.