Modeling Gene Regulatory Activity and the Effects of DNA Sequence Variants

Abstract

Cis-regulatory elements (CREs) at non-coding regions control gene expression. Advances in chromatin accessibility mapping, transcription factor binding profiling, functional reporter assays, CRISPR perturbation screens, and chromatin interaction technologies have collectively enabled large-scale characterization of regulatory elements across the genome. However, interpreting how signals from these assays relate to one another and how the regulatory activity captured in functional assay connects with epigenomic and chromatin architecture features remains an important challenge. In this work, I addressed these challenges by systematically comparing functional readouts across high-throughput reporter assays and genome-wide CRISPR-based perturbation screens. I then integrated their significant regulatory activity with transcription factor binding profiles and chromatin interaction data. Next, I examined how regulatory variations at genomic sequence alter the regulatory activity at the motif-level. Through these integrative analyses, the results provide a comprehensive examination for understanding how different experimental assays capture distinct aspects of cis-regulatory activity and how these different genomic features and TF motifs together contribute to explaining the regulatory mechanism of cis-regulatory elements.