Determining genotype-phenotype relationships for gene variants that increase susceptibility to autism spectrum disorder (ASD) is critical for understanding underlying mechanisms and identifying therapeutic targets. One of the main challenges is to interpret variant function with clinical predictive value for causal ASD genes. For the majority of rare missense variants this requires experimental assays. Yet developing cost-effective assays that provide reliable, well-calibrated assessments of disease-relevant variant function remains a persistent challenge.

In a previous SFARI-funded project, the labs of Christopher Loewen, Douglas Allan and Paul Pavlidis developed computational and experimental approaches to implement scalable, clinically predictive variant functionalization methods^1-4. Their approach tested human variant function in intact, genetically tractable organisms to reliably map variant function within disease-relevant pathways in a controlled, reproducible manner.

In the current project, the researchers plan to focus on molecular network perturbance by ASD variants. In any genetic disease, including ASD, gene variants can be considered perturbants of molecular networks. In this project, Pavlidis’s lab plans to identify genes and variants of interest. Allan and Loewen’s labs aim to establish variant functional assays, calibrated with clinically and biochemically characterized variants, in the genetic models S. cerevisiae (budding yeast) and D. melanogaster (fly). This allows exploitation of genetic interaction methods in intact, in vivo genetic systems to leverage these models’ inherent pathway integrity and experimental reproducibility so as to discreetly map variant function within disease molecular networks. Jesse Chao’s lab plans to examine variant stability and localization in human cell culture models.

This combined approach is expected to help define ASD variant function at an unprecedented scale, regarding the number of variants tackled, the speed of assessment and the clinical utility of the data generated. This is critical to achieve much needed progress in defining ASD risk variants and their disease pathways.

1. Post K.L. et al. Nat. Commun. 11, 2073 (2020) PubMed
2. Ganguly P. et al. PLoS Genet. 17, e1009774 (2021) PubMed
3. Young B.P. et al. Dis. Model Mech. 13, dmm044560 (2020) PubMed
4. Chao J.T. et al. Cancer Res. 80, 2775-2789 (2020) PubMed