In 2003, Duke University researchers published a landmark paper suggesting that a single gene helps determine a person’s risk of depression in response to divorce, job loss or another upset. This provided comfort, and an excuse, to some patients, the New York Times reports, who could point to their genes as a cause for depression. Yet on Tuesday, a group of prominent researchers reported this finding—one of the most celebrated in modern psychiatry—hasn’t held up to scientific scrutiny. The new report agrees that interactions between genes and life experience are likely important, but suggests that confirming it with any precision will be very difficult, and that the original finding may have been due to chance. This is bound to stir up debate within the field of psychiatry, as the genetics of illnesses like schizophrenia and bipolar disorder haven’t yet been pinned down. Since the 2003 finding, researchers have tried to replicate it more than a dozen times, with mixed success. In the new paper, published Wednesday in the Journal of the American Medical Association, a coalition of researchers reanalyzed the data in 14 studies, and found “no evidence of an association between the serotonin gene and the risk of depression,” no matter what the subject’s life experience was. On the other hand, a major stressful event was shown to raise the risk of depression by 40 per cent.