Title: “Advancing variant effect prediction with protein language models”
Assistant Professor, Dept. of Epidemiology & Biostatistics,
University of California, San Francisco
Protein altering genetic variants can have profound impacts on cellular/organismal phenotypes, and being able to quantify their effects is fundamental to research endeavors spanning clinical, population, evolutionary, and statistical genetics. Variant effect prediction (VEP) has been an ongoing challenge for more than 20 years, and despite numerous advances, computational methods are still not considered reliable enough for most applications. Our group and others have recently demonstrated the potential of protein language models for VEP, showing marked improvements in prediction accuracy. In this talk we will discuss protein language models as one of the most promising approaches to determine the clinical and biological consequences of genetic variants, overview the state-of-the-art and potential limitations, and explore new opportunities to further extend the utility and performance of variant effect prediction methods.
- Brandes et al. Genome-wide prediction of disease variant effects with a deep protein language model. Nat Genet 55, 1512–1522 (2023). https://www.nature.com/articles/s41588-023-01465-0
- PLM Review: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8050421/
- VEP Review: https://pubmed.ncbi.nlm.nih.gov/35736673/