Exploration and Visualization Methods for Chromatin Interaction Data

The novelty and sophistication of biological data present numerous challenges for data analysis. Among these challenges is the basic issue of how to interpret a biological dataset, particularly when the data in question is not well-standardized or fully understood, such as in the case of high-throughput chromatin conformation capture or Hi-C. Using Hi-C contact lists from publicly available databases as well as supplemental data, we demonstrate the utility of a filter-based approach in generating comprehensible graphs for Hi-C data that can be used to identify features of particular interest. We use our processing and visualization framework to produce chromatin interaction graphs specifically for cliques to illuminate the use of our filters to identify previously indistinguishable features in our large datasets and comprehensively validate their functionality. We suggest how this approach can be generalized to other visualizations of genomics data.