Functional assessment of de novo missense variants associated with Autism Spectrum Disorders through an overexpression-based screen in Drosophila

Whole-exome sequencing (WES) is becoming less expensive and more widespread as research and diagnostic tools for human diseases. WES identifies hundreds of rare variants in an individual’s genome that needs to be interpreted by human geneticists and clinicians. However, current in silico tools are insufficient to predict the pathogenicity of many missense variants. This becomes especially problematic when undertaking large sequencing efforts to uncover rare variants that contribute to the pathology of relatively common conditions. For example, numerous recent efforts have uncovered rare gene variants associated with intellectual/developmental disabilities and Autism Spectrum Disorder (ASD), but thousands of de novo variants are missense variants that are of unknown significance. Here, we utilize an overexpression-based screen in Drosophila to investigate several such de novo missense variants identified in the Simons Simplex Collection (SSC), a large cohort of over 2,500 ASD simplex families. Using this method, we prioritized 78 genes and generated UAS-human cDNA transgenic lines that allow exogenous expression of reference or variant alleles in flies. From these experiments we discovered 22 gene variants (28%) which had a scorable phenotypic difference between overexpression of the variant and reference allele. These data strongly suggest these variants have a consequence on gene function and may contribute to the pathogenicity of ASD.