(Virtual) LIN Seminar: “Functional variomics and deep learning networks: connecting genotype to phenotype in neurological disorders” by Stephen Yi (University of Texas at Austin)

Please join us on Oct. 12, 2023 for a virtual LIN Seminar featuring "Functional variomics and deep learning networks: connecting genotype to phenotype in neurological disorders" by Dr. Stephen Yi (University of Texas at Austin).

Website

Host: Rui Gao

Abstract:

It has been increasingly appreciated that each mutation of a gene could have a subtle but unique effect on protein function or network rewiring, contributing to diverse phenotypic consequences in neurological disorders. Genomic variants and regulatory molecules interact with each other in cellular networks. Such information is crucial to understanding the complex pleiotropic effect of liable genes and provides a possible link between genotype and phenotype. Although previous studies identified many genes and mutations associated with neurological disorders, they were mostly limited owing to patient sample size and selection bias. There is a lack of comprehensive research aimed to identify and prioritize risk mutations systematically. In this talk, I will present our recent work to study the diverse functional consequences of mutations at the base-pair resolution by our functional variomics technology. To address this challenge, we construct a large-scale mutation and co-mutation framework, and devise deep learning models configured with fully connected layers that are capable of predicting cognitive impairment in patients.

Discovering effective therapies is difficult for neurological disorders in that disease progression is often associated with complex and interactive mechanisms. Toward this end, we develop a novel deep learning framework based on multiple features including drug chemical structures, target gene features, and interaction profiles between drugs and genes, to identify potential repurposed drug therapies for neurodegenerative disease. More importantly, our framework is broadly applicable and may generalize to identifying potential drug combinations in other diseases.