Tracing Your Ancestry

Continued from page 1

These tests "can tell you something about lots of people, but they are not exhaustive," says Mark Shriver, associate professor of biological anthropology at Pennsylvania State University and a consultant with DNA Print Genomics, a Sarasota, FL, company that provides genetic tracing services. Some major groups, including South and Central Asian ones, are not as well represented in the databases, Shriver says. In addition, European groups need to be broken down further and Africa is even more understudied. "We are still limited by the databases," says Jason Eshleman, senior research scientist at the company. "The world is a big place with a lot of people and when you split it into four populations, you find out there are a lot of populations we don't know much about."

In both lineage and admixture testing, the larger the databases used to compare with a client's DNA, the more accurate the results are likely to be. Even so, large databases may not cover everyone. For example, if a client has a close database match with a person living in western Africa, does this mean that region is the most likely ancestral origin of his or her maternal or paternal lineage -- or could that person be a closer genetic match to people from elsewhere, but who're not yet included in any database?

"We take pride in curating the largest global and proofread academic database offered by any service, more than 35,000 individuals," Forster explains. "But even so, if we take a sample of, say, 100 Sicilians, the database will indeed map about 95 percent of them to Europe, but it will also identify the other five percent as being nonEuropean." This could be an error or it could reflect prehistoric migrations in and out of Sicily. "Clearly, the better the sampling of a continent, the lower this kind of error rate becomes, and improved sampling is something we are doing continuously as part of our research," says Forster.

With admixture testing, if results indicate that someone is, for instance, five percent East Asian, it could be true or it could be an error. If that person has other evidence, however, that he or she has a grandparent of East Asian descent, it makes a stronger case for the results being accurate. "These tests should be interpreted in light of other information," Shriver says. "You have to take the totality of information."

Beyond questions about a test's accuracy there are issues about how such findings might be used. Some social scientists point out that a person's ethnic identity is far more than genetics. While using DNA testing to learn about your genetic family tree is one thing, hanging one's entire identity on a genetic test is problematic, says Paul Brodwin, associate professor in the department of anthropology at the University of Wisconsin-Milwaukee. "The actual knowledge is disproportionate to what people do with it," he says. "What people do with this knowledge is weave a story of their identity. Some seize upon a tiny piece of evidence to say, 'ah-ha, now I know where I come from and who I am'."

Standard genealogical sleuthing -- combing through historical documents, reading books, talking to older relatives, and remembering oral stories passed down -- may give a person more information about their ancestry than a DNA test, says Brodwin.

So what is the value to me of finding out where in Africa some of my ancestors may have come from? Is it worth the $200 or more that tests might cost? Sure, because it's important for me to know anything about my African ancestors. But it's also important not to get carried away with the information. If a test suggests that I have lineage in Ghana, I'm not going to all of sudden start calling myself a Ghanaian. Nevertheless, I will have a small bit of information about some of my ancestors. And that's well worth knowing.