While hundreds of autism risk genes have been identified through analysis of de novo variants in probands with autism, understanding of the molecular mechanisms underlying these genetic contributions remains limited. Missense variants offer unique opportunities for mechanistic insights when mapped onto three-dimensional protein structures. This project focuses on protein-protein interaction interfaces.

Hilary Finucane and Mark Daly will systematically identify specific protein interaction interfaces enriched for de novo missense mutations in autism probands, leveraging unprecedented sample sizes now available through aggregation of trio sequencing data sets. First, they will map genetic mutations to protein interaction interface residues using multiple databases of experimentally determined or computationally predicted interaction interfaces. For each interface, we will test for enrichment of de novo mutations in autism probands relative to ultra-rare untransmitted alleles from the same trios. Second, they will use computational tools to estimate whether de novo variants systematically weaken protein interactions compared to ultra-rare untransmitted alleles.

This work will provide the first systematic identification of specific protein interaction interfaces disrupted by de novo variants in autism, moving beyond gene-level associations to pinpoint the binding events affected by disease variants. The identified interfaces will serve as potential targets for therapeutic development using molecular glues to restore weakened protein interactions. The framework with provide a generalizable method for identifying therapeutically actionable protein interfaces across neurodevelopmental disorders.