National Human Genome Research Institute 5R01HG012367-04

Sequence-based Machine Learning for Inference of Dynamic Cell State Gene Network Models

This award funds research to develop machine learning models that predict how genetic variations affect gene regulatory networks during cell state transitions, using high-resolution genomic data and CRISPRi validation, to better understand development and disease mechanisms.

Description

The project aims to develop computational methods to infer quantitative models of combinatorial interactions of transcription factors and enhancers by training on temporally-resolved measurements of gene activity, enhancer activity, and core cell fate-regulating transcription factor activity across cell state transitions in early human development. The research involves generating high-resolution genomic data and validating gene network predictions with CRISPRi in a native genomic context, focusing initially on embryonic-stem-cell to definitive-endoderm system and generalizing to larger regulatory datasets. The goal is to enable a quantitative understanding of how altered regulatory element activity affects gene regulatory networks and cell state transitions relevant to development and disease.

View original record