PhD Dissertation Defense: Maya Sheth

Title: Mapping Enhancer-Gene Regulatory Interactions from Single-Cell Data

Abstract: Mapping enhancers and their target genes in specific cell types is essential for understanding gene regulation and the impact of human genetic variation on disease. However, accurately predicting enhancer-gene regulatory interactions from single-cell datasets has been challenging. Here, we introduce a new family of classification models, scE2G, to predict enhancer-gene regulation. These models use features from single-cell ATAC-seq or paired RNA/ATAC-seq multiomic data and are trained on a CRISPR perturbation dataset including >10,000 experimentally tested element-gene pairs. To validate scE2G, we benchmark the models against CRISPR perturbations, fine-mapped eQTLs, and GWAS variant-gene associations and demonstrate state-of-the-art predictive performance across multiple cell types and categories of perturbations. Using scE2G, we build regulatory maps in heterogeneous tissues and show how they can be integrated with complementary models and datasets (i) to link noncoding variants to causal genes and cell types and (ii) to perform systems-level analyses identifying disease-relevant cell types and gene programs. Together, these results demonstrate that scE2G enables systematic, cell-type-resolved enhancer-gene mapping and provides a foundation for interpreting noncoding variation and dissecting the cellular basis of disease.

Please contact Sofia Rakicevic-More for the Zoom link.