Predicting the effects of missense variants using protein structure and sequence

The explosion in sequencing of human genomes has revealed millions of missense variants that change protein sequences, yet we only understand the molecular and phenotypic consequences of a minute fraction of these variants. Equally important, we lack the mechanistic understanding necessary to develop therapeutic strategies for diseases caused by missense variants. In this project we aim to develop computational methods to predict which variants cause disease and why. We will combine recent developments in machine learning for protein structure and sequence in a unique model that integrates the two sources of information. The model will help predict whether a specific variant observed in a patient is likely to cause disease, and the molecular mechanisms by which this happens.