Race is less valuable than using genetic variants to infer ancestry when it comes to biomedical research.

That's because the use of genetic polymorphisms, as the variants are called, sorts people into categories that are much more precise than classifying them by race, said Dr. Michael J. Bamshad, associate professor of pediatrics at the University of Utah School of Medicine. Although race can provide some useful information, it can be inadequate and even misleading when it comes to finding genetic links to disease.

Bamshad and science writer Steve E. Olson review the body of research available on the topic for the cover story of the December issue of Scientific American, titled "Does Race Exist?"

"Race is not as useful as other ways we have to sort people," said Bamshad. "There is information in race, but there is information in many other labels used to define populations. And we have better ways to sort people into groups."

Researchers would learn more by sorting people using genetic polymorphisms, he said, identifying both traits in common in groups and traits that are different.

"One of the ultimate goals is to be able to make inferences or predictions about health-related susceptibilities in individuals. To do that, we need to understand how genetic variation is distributed among different human groups," Bamshad said.

For example, people from Africa, Australian aborigines and folks from southern India bear a superficial resemblance because of skin pigmentation, but they are genetically quite different. The skin color is a trait that developed to protect them from the sun blazing overhead. And racial definitions vary worldwide. Someone "black" in the United States might be "white" in Brazil and "colored" in South Africa, where there are others classified as "black" and "white."

On the other hand, two groups who are similar genetically might have been exposed to different selective forces, so they appear to be more different than they are.

"Common notions of race do not always reflect a person's genetic background," they wrote.

The goal is to find biological relationships that will allow researchers to sort people into groups that offer telling information about them. Sorting is important to geneticists looking for underlying causes of disease. It can also be important in forensics, evolution, even understanding history, Bamshad said. Genetic markers are more accurate and informative.

Ideally, Bamshad said, a genetic polymorphism for sorting people into groups would always be found in one group and never in another. It doesn't happen. Instead, "there might be two types of markers, A and B, in one group and the same in another group. But A is found in 90 percent of those in the first group, but in only 40 percent of those in group 2. Using many polymorphisms like this, you can sort people into groups 1 and 2."

Polymorphisms play a role in individual susceptibility to AIDS, for instance. Some people, the article states, have a small deletion in both copies of a gene that encodes a particular cell-surface receptor, so they fail to produce those receptors on the surface of their cells. Most strains of the virus that causes AIDS bind to that receptor, so people who don't have it are resistant to the HIV-1 infection. That polymorphism is found almost solely in groups from northeastern Europe.

Other variations in that same receptor allow the infection but change the rate at which it leads to AIDS and death.

E-MAIL: lois@desnews.com