CLOSEST RELATIVE: Two groups sequenced parts of the Neandertal genome and estimated that humans and Neandertals (above, a skull of the latter hominid) diverged between 500,000 and 700,000 years ago
© DK LIMITED/CORBIS

A short, fossilized femur from a 38-year-old Neandertal, which sat untouched in a museum in Zagreb, Croatia, could lead to the first full genome sequence of Homo sapiens's closest relative and help scientists understand what is special about humans, say teams that published analyses of two partial sequences of Neandertal DNA in this week's issues of Science and Nature. "We're at the dawn of Neandertal genomics," proclaims Edward Rubin, genomics division director at the Lawrence Berkeley National Laboratory and lead author of the paper in Science. "In the next few years, using advances in DNA sequencing that's occurring, there's no question that we're going to have a Neandertal genome."

Working off samples from the same 38,000-year-old bone found in Vindija Cave in Croatia, the two groups--Rubin's and the coalition behind the Nature paper, led by Svante Pääbo, director of the department of evolutionary anthropology at the Max Planck Institute in Leipzig, Germany--used different methods to obtain their sequences. Rubin's used a metagenomic approach, incorporating fragments of Neandertal DNA into loops called plasmids, amplifying them in bacteria and then using known sequences of human DNA to isolate strands for sequencing. Pääbo's group used a technique developed by Branford, Conn.-based biotech company 454 Life Sciences called pyrosequencing, which takes advantage of the fact that ancient DNA is often fragmented. The first step breaks the DNA apart. The pieces are then sequenced directly, and researchers reassemble them by mapping them to similar sections in the human genome.

Rubin's method garnered 65,000 base pairs for sequencing, while Pääbo--who was also a co-author of the Science paper--was able to analyze one million base pairs. Each of these figures is a drop in the bucket considering the human genome is 3.2 billion base pairs long. Comparing the similar sequences of Neandertal and human DNA, Rubin's group determined that the two genomes are at least 99.5 percent identical. (For reference, the chimpanzee, humans' closest living relative, has a genome that is nearly 99 percent the same.) The two groups both estimated the time when the ancestral line that would become Neandertals and the line that would lead to modern humans diverged from a common ancestor: Rubin's group determined it took place roughly 706,000 years ago (with an error range of 468,000 to 1.015 million years); Pääbo's number is 516,000 years, with error ranging between 465,000 and 569,000 years. Because the groups' error bars overlap, both researchers say that they are statistically equivalent. "As we generate more sequence, [the divergence estimates] could come together," says Rubin, whose team also determined the point when Neandertal and human ancestors stopped interbreeding: about 370,000 years ago.

According to Rubin, the sequences provide the beginnings of a "DNA time machine" that will help update anthropological inferences about human and Neandertal populations. Among the lingering questions is whether the two populations intermixed after humans migrated out of Africa and encountered Neandertals in Europe 30,000 to 40,000 years ago. (Just this month two studies, from Washington University in St. Louis and the University of Chicago, suggested that indirect evidence from human DNA indicates intermingling occurred.) Both Rubin and Pääbo report finding no evidence of mixing. "We don't exclude it," Rubin says. "Clearly, as we go further into the future and read more, we may see evidence of that." Erik Trinkaus, a physical anthropologist and lead author of the Washington University study, believes Rubin's and Pääbo's results do not preclude his hypothesis. He says that there are two different questions regarding population mixing: Did it occur 40,000 years ago? And, do 21st-century Europeans carry distinctively Neandertal genes? "They are attempting to answer the second question and make a statistical inference back to the first question," Trinkaus explains.