Genome study gets insights in Africa

For all the recent revelations about the human genome, one area that has been largely overlooked is the place where humanity began: Africa.

University of Pennsylvania geneticist Sarah A. Tishkoff filled in a lot of gaps yesterday, with new findings that underscore Africa's status as the most genetically diverse place on Earth.

After years spent bumping along in a Land Rover to gather blood samples from remote tribes, she and her colleagues completed what is by far the biggest-yet study of African genetics.

The results, published in the online edition of the journal Science, offer insight into the migrations of ancient peoples as well as the makeup of modern African Americans.

The study also will provide a basis for future inquiry into which drugs work best on people of various ethnicities, Tishkoff said.

"We don't want to see African populations get left behind in this genomics revolution," said Tishkoff, who led a large scientific team from the United States, Europe and Africa itself.

The researchers analyzed samples taken from 2,432 Africans in 113 distinct populations or tribes, some of whom had never seen light-skinned people before. The scientists also compared the African data with the DNA of 98 African Americans from Chicago, Pittsburgh, Baltimore and North Carolina.

The analysis found that on average, those African Americans derived 71 percent of their ancestry from West Africa, conforming to historical accounts of the slave trade. About eight percent came from elsewhere in Africa. On average, 13 percent of their makeup was European, ranging from zero in some to more than 40 percent in others.

The African Americans in the study revealed "very little" evidence of American Indian ancestry, Tishkoff said.

While there are genealogy services that claim they can pinpoint for black Americans the African tribe from which they are descended, Tishkoff said that even with her extensive data, it is hard to draw such fine distinctions.

As for Africa itself, the genetic data confirmed earlier indications that the continent is a veritable rainbow of genetic variation when compared with the rest of the world.

That's because those who migrated out of Africa 50,000 years ago to Europe and beyond were only a small genetic subset of the African population.

The "landmark study" should erase any vestiges of colonial-era thinking that Africa is one unit, said Roy King, a Stanford University associate professor of psychiatry and behavioral sciences.

"No longer can we see Africa as just this homogeneous population," said King, who has studied African migration patterns but was not part of the new paper.

Though the researchers needed to be mindful of political unrest and infectious disease, a bigger challenge was simply the logistics of doing science in a distant land.

Sometimes there was no electricity to operate their portable centrifuge - needed to separate out white blood cells from which DNA was later extracted. So they had to plug the device into the battery of their Land Rover, said Tishkoff, who joined the Penn faculty in January 2008.

There was no refrigeration, so they had to add a chemical buffer to their samples to stabilize the DNA.

The researchers analyzed 1,327 genetic markers from each of their subjects. Those markers had nothing to do with any particular disease, rather they are well-known markers of human identity, Tishkoff said.

Yet they should prove valuable in future studies of disease, she said. In epidemiological studies where sick people are compared with controls, for example, it is important to select controls with similar genetic makeup to avoid skewing the results.

The anthropological information in the study may be of more immediate value.

In many cases the new genetic data have already enabled the team to link diverse African groups who migrated within the continent, said coauthor Christopher Ehret, a historian at the University of California, Los Angeles. Some share cultural practices and similar languages, though a few do not.

For example, the genetic data suggest that the pygmies of central Africa and the Hadza and Sandawe of eastern Africa were once part of a larger population of hunter-gatherers, the authors wrote.

The Hadza and Sandawe speak similar languages, characterized by "clicks." The pygmies do not, but may have done so in the past, given the shared genetic history, the authors wrote.

In addition to seeking approval from the African governments, team members said they took pains to explain their research to village elders and to individuals.

One of the coauthors, Jibril Hirbo, proved especially helpful in that regard. The graduate student in human evolutionary genetics is from northern Kenya and speaks Burji, Borana and Swahili, as well as some Somali.

Some villagers opted not to participate, a few because of a mythical belief that giving blood would weaken them in some way, Hirbo said.

But many were intrigued by the thought of learning about their history, said Hirbo, who is affiliated with Penn and the University of Maryland.

"If you tell them there is an imprint in their blood that may show how you're connected to different tribes," Hirbo said, "they can relate to that."