"We have various lines of evidence that there was more than one migration," said Dr. Andres Ruiz-Linares, a professor of human genetics at University College London and senior author of a report on the findings that was published Wednesday by the journal Nature.
The discovery was made possible by the sheer volume of genetic material the team was able to assemble and analyze, he said.
Ruiz-Linares and colleagues around the world analyzed DNA samples, primarily from blood, taken from hundreds of modern-day Native Americans and other indigenous people representing 52 distinct populations. These included Inuits of east and west Greenland, Canadian groups including the Algonquin and the Ojibwa, and a larger variety of people spanning the southern regions of the Americas from Mexico to Peru.
Investigating patterns in more than 350,000 gene variants, the scientists determined that most of the groups they studied did indeed descend from an original "First American" population.
However, they also saw that Eskimo-Aleut populations of the Arctic inherited almost half of their DNA from a second ancestral group, and the Na-Dene-speaking Chipewyans, from Canada, got about 10% of theirs from yet another group.
The results lined up nicely with a controversial model for the colonization of the Americas that was proposed in 1986 by Stanford University anthropological linguist Joseph Greenberg, Ruiz-Linares said.
By examining similarities between many native languages, Greenberg argued that the Americas were populated through three distinct migrations. But his hypothesis was widely rejected by researchers who didn't agree with his classification of languages.
In the years since, genetic studies seemed to bolster the single-migration scenario, Ruiz-Linares said. But most of those studies examined very limited sets of genetic data, such as variations on the Y chromosome or in mitochondrial DNA, which contains only a dozen genes and passes virtually unchanged from mother to child.
The new study employed computers to scan the entire genomes of all the volunteers and look for small genetic variants known as single-nucleotide polymorphisms, or SNPs. This finer level of detail revealed more subtle differences among populations.
"This makes it possible to ask and answer questions we couldn't even look at before," said study coauthor David Reich, a professor of genetics at the Harvard Medical School in Boston.
For instance, the analysis revealed that members of the separate migratory waves eventually mated with one another after arriving in the Americas.
Ripan Malhi, a molecular anthropologist at theUniversity of Illinois at Urbana-Champaignwho was not involved in the research, called the work "an important study" that marked "a big step forward" in using genomics to better understand the population history of Native Americans and other indigenous groups.
At the same time, he and others noted a key limitation in the research: The scientists had no samples from people living in the contiguous United States and only a few from Canada.
Deborah Bolnick, a genetic anthropologist at the University of Texas in Austin who also wasn't involved in the research, said that more samples from North America might provide evidence for even more migratory waves. "It's hard to tell if we're looking at three or many more," she said.
Ruiz-Linares, who was born in Colombia and began his population studies there in the 1990s, agreed that the samples included a disproportionate number from Central and South America because that's where his collaborators happen to be located.
By Eryn Brown