Leveraging chromatin modules to explore the impact of regulatory variation on disease

Human genome is complex and hierarchically organized within the nucleus, where cis-regulatory elements (CREs) frequently interact within topologically associated domains to regulate gene expression. While genomic technologies and computational approaches have enabled high-throughput mapping of CREs and their interactions in 3D, their precise role in gene regulation remains unclear. In this thesis we first review and unify concepts explaining similar observations of covariable CREs under the term "chromatin modules" (CMs). We also review the methods used to identify CMs and summarize potential mechanisms driving CRE interactions within these modules. In the subsequent chapter, we quantitatively characterize CMs, starting with an evaluation of existing CM mapping approaches, followed by an analysis of covariable CREs across different cell types. We demonstrate cell type specificity in gene regulatory variation at the level of CM-embedded genes, as well as the enrichment of cell type-specific transcription factor binding sites at CREs within CMs. Using high-throughput epigenome data in a range of genetic backgrounds, we show how CMs can help resolve the molecular mechanisms underlying cis-regulatory coordination and aid in the fine-mapping of disease- and trait-associated genetic variants. Finally, we discuss how CMs can be integrated into the current model of the functional 3D genome and outline the challenges in this area. Together, this work unifies complementary insights into CRE cooperation under the concept of "chromatin modules", provides a basis for future functional characterization of CRE interactions, and highlights the role of non-coding genetic variants in disease onset and progression.