Deep learning detects mutations behind autism

By Steven Schultz

Using artificial intelligence, a Princeton-led team has decoded the impact of a new class of mutations in people with autism. These mutations are not in actual genes but instead lie in the 99% of the genome that regulates the genes. Published May 27, 2019, in the journal Nature Genetics, the study sorted among 120,000 mutations in 1,790 families with more than one child in which one child has autism spectrum disorder but the others do not. The team applied an AI technique called deep learning to discover patterns that are otherwise impossible to find. “This method provides a framework for doing this analysis with any disease,” said Olga Troyanskaya, professor of computer science and the Lewis-Sigler Institute for Integrative Genomics. The work was funded by the National Institutes of Health and the Simons Foundation