The spatial organization of chromatin in the nucleus has been implicated in many aspects of regulated gene expression. Maps of high frequency interactions between different segments of chromatin have revealed Topologically Associating Domains (TADs), within which most of the regulatory interactions are thought to occur. Recent studies have shown that TADs are not homogeneous structural units, but rather they appear to be organized into a hierarchy. However, precise identification of hierarchical TAD structures remains a challenge. We present OnTAD, an Optimized Nested TAD caller from Hi-C data, to identify hierarchical TADs. Compared to existing methods, OnTAD has significantly improved accuracy and running speed. Results from OnTAD reveal new biological insights on the role of different TAD levels, boundary usage in gene regulation, the loop extrusion model, and compartmental domains. The software and documentation for OnTAD are available at: 

Hierarchical Domain Structure Reveals the Divergence of Activity among TADs and Boundaries