Summarizing internal dynamics boosts differential analysis and functional interpretation of super enhancers release_7kltakpofjbb3fdv3znkjbunka

Abstract

Super enhancers (SEs) are broad enhancer domains usually containing multiple constitute enhancers with significantly elevated activities. The constitute enhancers work together through chromatin looping to build up distinct regulatory properties of SEs. Aberrant SE activities, which are critical to understand disease mechanisms, could be raised by the alterations of one or more of their constitute enhancers. However, the state-of-art binary strategy in calling differential SEs only relies on overall activity changes, neglecting the local differences of constitute enhancers within SEs. We propose a computational method to identify differential SEs by accounting for the combinatorial effects of constitute enhancers weighted with their activities and locations (internal dynamics). In addition to overall changes, our method finds four novel types of differential SEs pointing to the structural differences within SEs. When applied to public datasets for six cancer cells, we demonstrate that different types of differential SEs complement each other with distinct sets of gene targets and varied degrees of regulatory impacts. More importantly, we found that some cell-specific genes are linked to SE structural differences specifically, suggesting improved sensitivity by our methods in identifying and interpreting differential SEs. Such improvements further lead to increased discernment of cell identifies.
In application/xml+jats format

Archived Files and Locations