Spatiotemporal and cell-type convergence to reveal autism neurobiology

Over the past few years, the number of genes associated with autism spectrum disorder (ASD) has increased substantially. More than 65 genes are strongly associated with autism, and current estimates suggest that hundreds of genes may play a role.

In the current project, Matthew State, Stephan Sanders, Kathryn Roeder and Bernie Devlin will further develop the TADA (Transmission And De novo Association) statistical framework¹ for gene discovery to include targeted sequencing data, genome sequencing data, structural variation, common variation and control data from more than 60,000 exome samples generated as part of the Exome Aggregation Consortium (ExAC)². This will allow the researchers to rank the degree of ASD association for all human genes.

Another algorithm that has proven valuable in identifying autism genes and biological pathways is DAWN (Detecting Association With Networks)³, developed by Roeder and Devlin. Whereas TADA identifies ASD genes based purely on genetic association, DAWN builds on the TADA results using other types of genomic data, such as RNA-Seq, to further inform gene discovery. The team plans to develop the DAWN method to incorporate multiple types of functional genomic data, including gene co-expression networks from tissues and single cells, transcription factor binding sites and markers of chromatin state.

Together, these methods will allow the researchers to detect ASD genes that might be missed by genetic association alone (for example, a short gene that is unlikely to have multiple observed de novo mutations in the current sized cohorts), thus increasing our understanding of the genetic landscape of autism.