Customers of 23andMe, a genetic testing service, contributed their data to a large-scale study of depression, allowing the study to succeed where earlier, smaller, and more conventional studies failed. The new study uncovered 17 genetic variations linked to depression at 15 genomic locations. In doing so, the new study has not only demonstrated the value of crowdsourced data, it has also established that depression is a brain disease with its own biology.
Results of the new study appeared August 1 in the journal Nature Genetics, in an article entitled, “Identification of 15 Genetic Loci Associated with Risk of Major Depression in Individuals of European Descent.” One of the article’s co-corresponding authors, Roy Perlis, M.D., a scientist affiliated with Massachusetts General Hospital, offered this perspective on the study’s significance:
“Identifying genes that affect risk for a disease is a first step toward understanding the disease biology itself, which gives us targets to aim for in developing new treatments. More generally, finding genes associated with depression should help make clear that this is a brain disease, which we hope will decrease the stigma still associated with these kinds of illnesses.”
In the Nature Genetics article, Dr. Perlis and colleagues (including colleagues from industry) described how they adopted a strategy of studying much larger samples than had been used in earlier studies, which had been conventional genome-wide association studies. This strategy, they noted, was designed to increase the odds of detecting weak genetic signals.
The conventional genome-wide approaches had failed toidentify reliably chromosomal sites associated with the illness in populations with European roots. Because depression is thought to be like fever—a common set of symptoms likely rooted in multiple causes—lumping together genetic data from people with different underlying illness processes likely washed out, or statistically diluted, subtle evidence of effects caused by risk genes.
“We used self-report data from 75,607 individuals reporting clinical diagnosis of depression and 231,747 individuals reporting no history of depression through 23andMe and carried out meta-analysis of these results with published MDD [major depressive disorder] genome-wide association study results,” wrote the authors of the Nature Genetics article. They added that particularly significant genetic loci identified in the meta-analysis were further analyzed in a replication data set consisting of 45,773 cases and 106,354 controls from 23andMe.
“A total of 17 independent SNPs from 15 regions reached genome-wide significance after joint analysis over all three data sets,” the authors continued. “Some of these loci were also implicated in genome-wide association studies of related psychiatric traits.”
While it is well known that depression can run in families, most previous genetic studies have been unable to identify variants influencing the risk for depression. One study did find two genomic regions that may contribute to disease risk in Chinese women, but those variants are extremely rare in other ethnic groups. Dr. Perlis and his colleagues note that the many different forms in which depression appears and affects patients imply that, as with other psychiatric disorders, it is probably influenced by many genes with subtle effects.
The researchers acknowledge that the genome sites identified still account for only a fraction of the risk for depression, but add that their results support the strategy of complementing more traditional methods with crowd-sourced data. They also speculate that their results could help guide efforts toward improved treatments for depression.
“The neurotransmitter-based models we are currently using to treat depression are more than 40 years old, and we really need new treatment targets. We hope that finding these genes will point us toward novel treatment strategies,” explained Dr. Perlis. “Another key takeaway from our study is that the traditional way of doing genetic studies is not the only way that works. Using existing large datasets or biobanks may be far more efficient and may be helpful for other psychiatric disorders, such as anxiety disorders, where traditional approaches also have not been successful.”