"You can take a single person's genome and learn an entire population's history from it," says David Reich, a geneticist at Harvard Medical School in Boston, Massachusetts, who was not involved in the study. "This is one of the dreams we've had as a community."

The analysis, published today in Nature1, suggests that descendants of the first humans to leave Africa dwindled to little more than 1,000 reproductively active individuals before rebounding. The study also suggests that, contrary to assumptions made from archaeological evidence, these early humans continued to breed with sub-Saharan Africans until as recently as 20,000 years ago.

Geneticists eager to plumb human history have traditionally compared DNA sequences from numerous people around the world to determine how different populations relate to one another and when they might have gone their separate ways. For instance, studies of DNA from maternally inherited cell structures called mitochondria established that all humans can trace their maternal lineage back to one woman — a mitochondrial Eve — who lived in Africa around 200,000 years ago2.

But, just as mitochondria can lead us back to a single woman, parts of a person's genome inherited from both their mother and father can also be followed back in time, with individual genes traced back to points before any mutations had developed, when just one version — a common ancestor — of that gene existed. Because of the way a person's maternal and paternal chromosomes shuffle together to create diversity in their sperm or egg cells, some parts of a person's genome inevitably share common ancestors more recently than other parts.

"Each little piece of the genome has its own unique bit of history and goes to a unique ancestor as you go further and further back," explains John Novembre, a population geneticist at the University of California, Los Angeles, who was not involved in the study. "As you look at different parts of the genome, you get access to different parts of history."