Investigating genetic burden between autistic individuals and their siblings

Autism is a genetically heterogeneous condition, with contributions from de novo variants, inherited rare variants, and polygenic common variation. Notably, some of the strongest genetic risk factors for autism, such as 16p11.2 deletions and duplications, are incompletely penetrant and observed in unaffected individuals. This phenomenon suggests that additional genetic or environmental modifiers influence whether carriers of large-impact variants develop autism. Under the liability threshold model, multiple genetic and environmental factors act together to determine whether an individual crosses the threshold for clinical diagnosis. However, the extent to which background genetic burden modifies the penetrance of large-effect autism-associated variants remains incompletely understood.

To address this question, Kaitlin Samocha and colleagues at Massachusetts General Hospital will leverage large-scale genomic data from family-based cohorts, including the Autism Sequencing Consortium, the Simons Simplex Collection and the SPARK initiative, focusing on greater than 7,600 sibling pairs discordant for an autism diagnosis. A key feature of this design is the comparison of affected and unaffected siblings who share a high-impact rare variant. Rare variant burden will be quantified using a unified severity framework that incorporates structural variants and protein-coding single nucleotide variants and insertions/deletions, weighted by gene-level intolerance to loss of-function variation. Common variant burden will be measured using polygenic scores derived from recent large-scale genome-wide association studies of autism, schizophrenia, and other cognitive traits.

By integrating rare and common genetic burden within sibling pairs, these analyses will test whether unaffected carriers of high-impact variants are protected by lower polygenic risk, and whether affected siblings show elevated aggregate rare variant load. These results will refine models of incomplete penetrance and advance understanding of how multiple classes of genetic variation jointly shape neurodevelopmental risk.