The project will focus on four existing epidemiological studies in which scientists have spent years tracking participants' medical information, such as blood pressure, medications, lifestyle, and nutrition. The people being studied encompass a broad swath of the United States' population, including African Americans, Hispanics, American Indians, and Asian and Pacific Islanders. That's especially important because most genome-wide association studies were done in people of European descent; other groups may carry that variant at different frequencies.

One of the groups under study is the population screened by the National Health and Nutrition Examination Survey, a survey done every year by the Centers for Disease Control, in Atlanta. "It's meant to be a snapshot of American health at the time the information is collected," says Dana Crawford, an assistant professor of molecular physiology and biophysics at Vanderbilt University, in Memphis, TN, who is leading one arm of the project. "We have data on pesticide exposure and occupational exposure--it's a huge opportunity to look at gene-environment interaction." (Genes have differing effects under different conditions. For example, some studies link pesticide exposure to Parkinson's disease, but this risk may only be realized in people who possess certain genetic risk factors.)

"I think this kind of effort will be useful because it will cast a wide net looking at the way results [from genome-wide association studies] can be used to improve human health," says Evans. "Arguably, much of the utility of this knowledge being generated by [these] studies will be at the population level and could have very useful effects on how we practice medicine from the perspective of public health."