Researchers use machine-learning approach to predict autism risk genes

Published on August 1, 2016 at 1:37 PM Princeton University and Simons Foundation researchers have developed a machine-learning approach that for the first time analyzes the entire human genome to predict which genes may cause autism spectrum disorder, raising the number of genes that could be linked to the disorder from 65 to 2,500. The findings will appear in the journal Nature Neuroscience. A PDF of the study is available on request. ASD is a complex neurodevelopment disorder with a strong genetic basis, but only about 65 autism genes out of an estimated 400 to 1,000 have been found through sequencing studies. Because of the how complex autism is, sequencing/genetics studies alone are severely underpowered to uncover the genetic basis of autism. So, the Princeton-led team developed a complementary machine-learning…


Link to Full Article: Researchers use machine-learning approach to predict autism risk genes