Guidelines Warn Against Racial Categories in Genetic Research


At the same time, environmental variables can have potent biological effects on people. Groups who live in neighborhoods with high air pollution, for example, will have higher rates of certain diseases. The authors of the new study recommend building these environmental influences into studies, rather than looking at genes in isolation.

The authors created a flowchart to help scientists design their research. Different kinds of research call for different kinds of experiments.

Scientists looking for individual mutations that cause severe diseases, for example, may not need to consider the ancestry of their subjects at all. The mutation may well cause the same disease regardless of the ancestry of the person carrying it.

Other scientists look at the DNA of many people to reconstruct the ancestry of human populations. Dr. Tishkoff and her colleagues recommend that scientists don’t throw their subjects together in groups that don’t reflect their deep history. A national label like “Tanzanian,” for example, describes only people who live in a country that gained independence in 1961.

Instead, the authors recommend that scientists identify their subjects with meaningful information, such as their local ethnic group or language. In some studies, they said, it may be appropriate to describe subjects by the percentage of ancestry they can trace to different populations. When they publish their findings, researchers should be transparent about the grouping decisions they made so that other researchers can revise the groupings based on new evidence, the guidelines said.

Joseph Graves, an evolutionary geneticist at North Carolina A&T University who was not involved in writing the report, said it offered scientists a path out of some of the fallacies that have hampered previous studies on human health and variation.

“The strengths are really to help researchers disentangle the social definitions from the biological definitions,” he said.

But he warned that simply putting out a report would not be enough. “We need to get out there and be talking to our colleagues,” he said. “The report can work, but it requires people to get behind it.”