Abstract
Recent research has established the significant role of newly arising (de novo) variation in neurodevelopmental disorders. Using aggregated data from 5603 families affected by ASD, intellectual disability (ID), or developmental delay (DD), we show that overall ~1/3 of de novo variants are independently observed as standing variation in the Exome Aggregation Consortium’s (ExAC) cohort of 60,706 adults without severe neurodevelopmental disease, and that these de novo variants do not contribute substantively to neurodevelopmental risk. We further use a loss-of-function intolerance metric derived from ExAC to identify a small subset of genes that contain the entire signal of associated de novo protein truncating variants (PTVs) in neurodevelopmental disorders (ASD: rate ratio (RR) = 3.24, P < 10−16; ID/DD: RR = 6.70, P < 10−38). These same genes carry a moderate excess of PTVs inherited by individuals with ASD from their unaffected parents (RR = 1.16, P = 0.00985), and a larger excess in 404 ASD cases and 3654 controls (OR = 2.66; P = 6.34 × 10−19), with again no association seen outside these genes. These findings illustrate the importance of population-based cohorts as reference samples for the interpretation of candidate pathogenic variants, even for the analysis of de novo variation.