NIH Reporter #5R21HG014015-02

From Text to Translation: Using Language Models to Resolve and Classify Variants

This award funds a project to develop computational models using large language models to classify genetic variants and improve disease risk prediction for clinical use.

Description

The project aims to improve clinical information for patients carrying genetic variants that may increase disease risk by developing computational approaches using large language models to predict pathogenicity of genetic variants. It involves re-using existing clinical reports, fine-tuning models, and updating variant classifications.

View original record