Did early humans interbreed with a ‘ghost’ population?

The evolutionary history of our species is an intricate web. Early populations expanded, migrated, or met with each other, sometimes branching out and at other times intermixing. Untangling this web has been a major scientific challenge, but in recent decades, researchers have refined models that use the genetic variation found in people today to peer back in time.

A problem they have run into, however, is that some genetic differences appear to be incredibly old—much older than the date at which our common ancestors are thought to have split into different populations. Inspired by recent evidence that Homo sapiens interbred with Neanderthals and Denisovans in Eurasia, some scientists have suggested that these old genetic variants may be explained if early humans in Africa also occasionally mated with another species.

No one has found physical evidence of these “ghost” hominins, however. There are no fossils, let alone DNA. And new research published in Nature provides an alternative explanation.

Based on knowledge of how fast our DNA changes from generation to generation, it is possible to estimate a time when the common ancestors of people carrying different gene variants still had the same ones. Models of the genetic evolution of our species so far have envisioned ancestral populations as the solid stem of a family tree that later split to create branches of separate populations. Any individuals within the stem would have been genetically similar, while ones on different branches would hardly mix anymore. Yet by modeling other scenarios, the new study suggests this stem may not have been as united as we thought.

“When we assume in our computer model that the stem population wasn’t quite as solid, but that parts of it would occasionally branch off and then later merge back together, we get a much better match with the genetic variation found in human populations today,” says population geneticist Aaron Ragsdale of the University of Wisconsin-Madison, lead author of the new study.

And geological clues can help explain what may have driven early groups of Homo sapiens apart or back together. “During the period we are interested in, roughly from a million to a 100,000 years ago, we know certain changes in the climate such as glacial cycles would have caused populations to expand or diverge into new areas in some periods, and contract or merge in others,” says study co-author and geneticist Brenna Henn of the University of California, Davis.

By assuming there was more exchange among the ancestors of Homo sapiens than previously thought, the model accounts for “these very old differences that previous models had struggled to explain without invoking ghosts,” Ragsdale says.

New genetic details

One intriguing episode emerging from the new model occurred around 120,000 years ago, at the end of a glacial period that caused a transition in parts of Africa from cold and arid to warm and humid conditions. Rising sea levels during this time may have driven people toward the interior of the continent.

“In this period, we see two branches of the human family tree merging to become the ancestors of today’s Khoe-San, a number of related but culturally distinct groups now confined to southern Africa that are genetically more diverse than everybody else on the planet combined,” Henn says.

The new analysis is the first to include genetic data from dozens of Nama people, a pastoralist Khoe-San group in Namibia that Henn has collaborated with for years to reconstruct its unique history. “It’s funny,” she recalls, “when I talk to some of the participants and tell them we find they have the highest genetic diversity, and that they’ve probably been isolated in southern Africa for many thousands of years, they look at me and say: Yeah, we know that.”

Many Nama were also hardly surprised to learn that about 15 percent of their genome today derives from European ancestors, as the family relations that gave rise to this are often quite recent. “More surprising was that they also have East African ancestors from around 2,000 years ago,” Henn says. “We are now working with the Nama to create an exhibit about this for the /Ai-/Ais Richtersveld Transfrontier Park.”

Around 100,000 years ago, the model suggests, a second merger between stems occurred that gave rise to the ancestors of West- and East-Africans, some descendants of whom later dispersed out of Africa to populate the other continents.

“This is in line with recent ideas from paleoanthropology that various populations within Africa have contributed ancestry to the group of Homo sapiens that left Africa,” Henn says. “It also shows we really need to be more specific instead of just speaking of African ancestry—the diversity is incredible.”

The new research supports the idea that “we are a species with many origins in Africa, with ancestry flowing from several populations, not just one,” says paleoanthropologist Eleanor Scerri of the Max Planck Institute for Geoanthropology in Germany, who was not part of the new study.

Squaring the fossil record

The new model could also change the interpretation of some surprising fossils with a mixture of archaic and modern features that have been found in various corners of Africa—sometimes interpreted as evidence of interbreeding with one or more ghost populations. A skull found near Kabwe in Zambia in 1921, for example, has quite a large braincase for such an old fossil, while other features, such as the heavy brow ridges and large face, look decidedly more archaic. The remains of a skull found near Iho Eleru in Nigeria in 1965 have small brow ridges and suggest a modern brain size, but the long and low shape of the braincase is more typical of archaic Homo.

The new model “further lowers the likelihood that the populations these specimens belonged to contributed to the surviving Homo sapiens lineage,” says paleoanthropologist Chris Stringer of the Natural History Museum in London, who contributed to the description of both fossils. “If there was any earlier mixing, all traces of it must have disappeared.”

Paleoanthropologist Jessica Thompson of Yale University, who was not part of the study team, points out that it would also be valuable for the models to incorporate ancient DNA recently found in Africa, some of which she has helped to uncover. Genome-wide DNA for six individuals from Tanzania, Malawi, and Zimbabwe that is between 5,000 and 20,000 years old was recently sequenced, for example, and could help create a more detailed picture of human’s evolutionary history. “People alive today may be quite different from those who lived in the same place in the past,” Thompson says.

Ragsdale and Henn agree it would be interesting to include ancient DNA from Africa in their analysis, but add that it is still very rare and covers only the most recent part of the period they’re interested in. The discovery of older DNA “would allow us to pin things down to certain locations, which we can’t do with our current data,” Ragsdale says. “So it would definitely add more detail. But I’m not sure it would really change the global picture.”

The new research presents an intricate image of our species’ origins in which early Homo sapiens were not confined to one area or population, and genetic variations that still exist may have evolved early on. It’s “kind of difficult to wrap your head around it,” Ragsdale admits.

It’s a common pattern in science: as our understanding grows, simple narratives fade and complexity accumulates. Whether Homo sapiens will prove clever enough to entirely elucidate the mystery of its own origin remains to be seen.