Clustering and Comparison of Hierarchies in the Spatial Organization of Chromatin

Abstract

Recent development of high-throughput 3C-based technologies made it possible to study spatial organization of chromatin in the nucleus and revealed new units of chromatin organization– Topologically Associating Domains (TADs). Each such domain is comprized of DNA regions actively contacting with each other while having fewer contact outside the domain. TADs are conserved between cell lines and even between related species. While there exist numerous bioinformatics tools for the TADs identification, the choice of a method and its adjustable set of parameters influences the number of the identified TADs and their characteristics. Besides, one more common disadvantage is that the most methods may leave gaps between the identified TADs. This study aimed to develop a robust and universal tool which would overcome these limitations. We have developed and implemented an algorithm, that builds a hierarchical tree on Hi-C contact matrices and methods to calculate the differences between such trees.