A brief, highly imprecise timeline: Around 600,000 years ago, a human-like species known as the Denisovans split off from our common ancestor. Some time after that, they met and mated with members of our species, enough to impart a handful of genes that have persisted to this day. Sometime after that, they vanished, until 2008, when researchers unearthed a pinky bone belonging to one of their kind, and officially welcomed them into the human family tree.

As John Wenz noted in The Atlantic in 2014, the Denisovans—named for Denisova, the cave in Siberia’s Altay mountains where the pinky bone was found—are “the first human cousin species identified with more than fossil records.” Everything scientists know about the species comes from the DNA they left behind, in the pinky bone, previously unidentified teeth, and in modern-day Melanesians, who can trace 2 to 4 percent of their genes back to the Denisovans.

The Melanesians are the people indigenous to the cluster of Pacific islands that includes Fiji, Papua New Guinea, the Solomon Islands, and Vanuatu, among others. They are the only group of people known to have Denisovan ancestry, though by no means the only ones to carry the genes of another species. Most modern humans outside of Africa have genomes that are roughly 2 percent Neanderthal (though, in a grand tradition of Atlantic Neanderthal coverage, I think I’m obliged to note here that my colleague James Fallows is at 5 percent).

Denisovans diverged from Neanderthals 400,000 years ago. Because the two species are more closely related to one another than they are to modern humans, it can be difficult to differentiate between the two when sequencing contemporary genomes, explained Josh Akey, a professor of genomic studies at the University of Washington. But in a study published today in the journal Science, Akey and his colleagues developed a new technique to more accurately determine which type of archaic DNA a person is carrying.

The researchers sequenced the genomes of 1,523 people from across the world, including 35 Melanesian people from Papua New Guinea. Their most striking finding: Across the board, there were vast stretches of the genome devoid of either type of archaic DNA—an absence that, the authors argue, offers a clue as to which specific genes helped us evolve into our present form.

The results “are potentially road maps in the genome that contribute to uniquely modern human phenotypes,” Akey said. “These depleted regions contain genes known to be involved in speech and language, in brain development, in all sorts of things that you might think a priori would be important to modern human evolution.”

The fact that those “depleted regions” were consistent across all the genomes, he said, means that the pattern was likely a result of natural selection over coincidence. “If it was just a random phenomenon of not seeing Neanderthal ancestry in those regions, we would expect to see some Denisovan sequences in those regions too,” Akey said. “But that’s not what we see. We see the consistent depletion of Neanderthal and Denisovan sequences, and that suggests that archaic sequence in general in those regions is incompatible with modern humans.”

Those genetic sequences—and lack thereof—may be as close as we can come to understanding who the Denisovans were, where they came from, and where they went before they disappeared. Without a fossil record, it’s unlikely that scientists can glean much about how they lived or even what they looked like. But they can still offer us a clues about our own history as a species.

“I’m hopeful that over the years, we can understand the biology of these regions more and learn about the genetic changes that are important in the evolution of modern humans,” he said. “Just thinking about, what are the reasons that modern humans survived and went on to be a population of 7 billion people, while these other archaic humans eventually became extinct?”