Last year, scientists released the results from the first genetic analysis of the STAR*D trial. They found that people with a specific variation in a receptor for the chemical messenger serotonin were more likely to respond to citalopram, an antidepressant that targets the serotonin system. Scientists are now analyzing the DNA of patients in the study who responded to different classes of antidepressants. "We want to determine ifgenetic makeup can tell us which medications are good or bad for individual patients," says A. John Rush, a psychiatrist at the University of Texas Southwestern Medical Center, in Dallas, who is leading the STAR*D trial.

A similar trial for bipolar disorder, called the STEP-BD (Systematic Treatment Enhancement Program for Bipolar Disorder) trial, wrapped up this month, and scientists are running similar genetic studies in search of the genetic roots of this disease. Both groups are focusing on specific regions of the genome that have previously been linked to the disorders, as well as using newer genomics technology--DNA microarrays that can analyze thousands of genetic variations in a single experiment--to scan the entire genome for clues. Scientists say that they expect to publish results within the next few months.

The findings might ultimately help psychiatrists redefine these complex disorders. For example, patients diagnosed with depression can have very different symptoms, and many symptoms are common to both depression and bipolar disorders. "We have diagnosis of mood disorders and other psychiatric disorders, but it's not entirely clear if genes are influencing these disorders per se, or if they influence specific symptoms that cut across diagnostic categories," says Smoller. "Genetic studies may tell us something about how we understand the relationships between different disorders."