Project Grant Award Summary

The University of Chicago received a $357,809 Project Grant from the National Institute of General Medical Sciences (NIGMS) under the Biomedical Research and Research Training program (CFDA 93.859), effective June 1, 2026 through May 31, 2031. This five-year research initiative focuses on developing next-generation sequence-based artificial intelligence (AI) frameworks for decoding the molecular mechanisms of genome regulation. The project delivers three primary research products: (1) explainable sequence models designed to discover simple regulatory rules across diverse genomic functions; (2) novel computational architectures that predict long-range regulatory interactions directly from DNA sequence and generate genome-wide regulatory interaction maps across multiple cell types; and (3) pioneering sequence-to-distribution generative models that capture the dynamic, single-molecule nature of chromatin regulation and reveal molecular cooperativity determinants.

This NIGMS-funded research addresses fundamental limitations in current deep learning approaches to genomic prediction by emphasizing biology-informed simplicity, explainability, and mechanistic insights. The project leverages the University of Chicago's established research capabilities in computational biology and biomedical sciences to advance understanding of how DNA sequences encode regulatory activities, ultimately contributing to improved predictions and deeper mechanistic insights into genome regulation that can inform future advances in disease diagnosis, treatment, and prevention.