Shared and distinct phenotypic profiles among neurodevelopmental disorder genes

Rare pathogenic variants in many genes contribute to neurodevelopmental disorders (NDDs), including intellectual disability and/or global developmental delay (ID), autism spectrum disorder (ASD), epilepsy (EP), and cerebral palsy (CP). These conditions frequently co-occur and share genetic etiologies, yet the broader phenotypic effects and the extent of shared versus distinct genetic influences remain unclear. Here, we adopt a cross-disorder framework to examine NDD genes across four diagnostic categories, characterize gene-associated phenotypic profiles, and identify convergent pathways that help refine how pathogenic variants in these genes shapes clinical outcomes. Using a discovery cohort of 8,973 probands with disease-causing variants in 263 NDD genes, we performed phenotype-based gene clustering and identified six distinct gene clusters. These clusters reveal structured patterns of genetic overlap, showing that subsets of NDD genes preferentially contribute to specific disorder combinations of ID, ASD, EP, and CP. The largest gene cluster was characterized by ID, whereas the other five included one enriched for ASD and ID, two for EP and ID and two for CP and ID, each with significantly differing frequencies. In an independent validation cohort of 19,704 probands, five of the six clusters were replicated. Gene Ontology enrichment analyses revealed distinct biological processes in each cluster, suggesting that coherent molecular mechanisms underlie the differing NDD diagnostic profiles. Together these findings demonstrate that NDD genes fall into coherent clusters that consistently map onto characteristic phenotype profiles, providing a framework to inform future therapeutic strategies and support early prognostication for individuals with pathogenic variants in NDD genes.