The DNA doesn't match modern humans or Neanderthals, two species that lived in that area around the same time - 30,000 to 50,000 years ago.
Instead, it suggests the Siberian species lineage split off from the branch leading to moderns and Neanderthals a million years ago, the researchers calculated. And they said that doesn't seem to match the history of human ancestors previously known from fossils.
So the Siberian species may be brand new, although the scientists cautioned that they're not ready to make that claim yet.
Other experts agreed that while the Siberian species may be new, the case is far from proven.
"We really don't know," said Ian Tattersall of the American Museum of Natural History in New York, who wasn't involved in the new research.
But "the human family tree has got a lot of branchings. It's entirely plausible there are a lot of branches out there we don't know about."
The discovery "is like many new finds," said Eric Delson of Lehman College of the City University of New York, who didn't participate in the new work. "You say, 'I think this is different, but I'm not sure.' And then you look for more material and you try to make better comparisons."
The researchers, who say the Siberian species is not a direct ancestor of modern-day people, hope further genetic analysis will show if it's a new species. Some experts are skeptical about whether such analysis will resolve that.
In any case, the finding emphasizes that quite unlike the present day, anatomically modern humans have often lived alongside their evolutionary relatives, one expert said.
"We weren't alone," said Todd Disotell of New York University, who was familiar with the new work. "When we became modern, we didn't instantly replace everybody. There were other guys running around who survived quite well until very, very recently."
Just last month, other researchers used DNA analysis to show the genetic diversity still present in residents of Africa, the cradle of the human race. And another project produced the first genome of an ancient human - a man who lived in Greenland some 4,000 years ago.
The new work, published online Wednesday by the journal Nature, is reported by Johannes Krause and Svante Paabo of the Max Planck Institute for Evolutionary Anthropology in Leipzig, Germany, and others.
They describe mapping DNA from what appeared to be a youngster's pinkie finger bone, which had been recovered in 2008 from Denisova Cave in Altai Mountains of southern Siberia. They showed how it differed from DNA of 54 modern-day people and six Neanderthals.
Their analysis indicated the Siberian species last shared a common ancestor with modern humans and Neanderthals about 1 million years ago. That in turn suggested there was a previously unrecognized migration out of Africa around that time, they said.
The work decoded the complete set of DNA from mitochondria, the power plants of cells. That's different from the better-known DNA that comes from cell nuclei and determines things like eye color. Paabo said the researchers are working to decode nuclear DNA from the Siberian species. That will reveal whether it was closely related to Neanderthals or today's humans, and answer questions like whether it interbred with Neanderthals or ancestors of modern-day people, he said.
Without a completed analysis of the nuclear DNA, "we are not saying this is a new species," Paabo said, although he said that's a likely possibility.
Rick Potts, director of the Smithsonian Institution's Human Origins Program, said the Siberian find might represent Homo heidelbergensis or Homo erectus. And even analysis of the Siberian species' nuclear DNA won't show if it's distinct from those ancestors, he said.
As for the study's suggestion of a migration out of Africa about a million years ago, Potts said there's already evidence of one or two migrations around that time.
The finger bone recovered from the Siberian species is not enough for a fossil-to-fossil comparison with other ancient species to show whether it's a new species, Delson said.
He suspects it might be a descendant of Homo erectus that's already documented in some fossil remains in northern Africa and Europe. Scientists are still trying to figure out how many species of the Homo grouping those bones represent and what name or names to attach to them, he said.
Disotell said the new creature could be an early version of Homo antecessor, a forerunner of Neanderthals and modern humans known from fossils in Spain. Or, he said, it could be a new species. In fact, the eventual decision could hinge mostly on the philosophical question of just how different a creature has to be to be declared a new species, he said.
Potts said that in the new work, "what we're seeing is a really, really interesting distant echo of the DNA history of human evolution.... This is an amazingly powerful technique that these guys have. This is going to be a growth industry in the study of human evolution."
On the Net: