TY - JOUR T1 - MrTADFinder: A network modularity based approach to identify topologically associating domains in multiple resolutions JF - bioRxiv DO - 10.1101/097345 SP - 097345 AU - Koon-Kiu Yan AU - Mark Gerstein Y1 - 2016/01/01 UR - http://biorxiv.org/content/early/2016/12/29/097345.abstract N2 - Genome-wide proximity ligation based assays such as Hi-C have revealed that eukaryotic genomes are organized into structural units called topologically associating domains (TADs). From a visual examination of the so-called chromosomal contact map, however, it is clear that the organization of the domains is not precisely defined. Instead, TADs exhibit various length scales, and in many cases nested organization can also be found. Here, by exploiting the resemblance between TADs in a chromosomal contact map and densely connected modules in a network, we formulate TAD identification as an optimization problem and propose an algorithm, MrTADFinder, to identify TADs from intra-chromosomal contact maps. MrTADFinder is based on the concept of modularity. A key component is to derive a background model for any given contact map, by numerically solving a set of matrix equations. The background model preserves the coverage of each genomic bin as well as the distance dependence of contact frequency for any pair of bins exhibited by the empirical map. Also, by introducing a tunable resolution parameter, MrTADFinder provides a self-consistent approach to identify TADs at different length scales, or resolutions. At a low resolution, larger TADs are found whereas, at a high resolution, smaller TADs are identified. We then apply MrTADFinder to identify TADs in various Hi-C datasets. The identified domains exhibit boundary signatures that are consistent with the earlier works. Moreover, by calling TADs at different resolutions, we observe that boundary signatures change with respect to the resolution, and different chromatin features may have different characteristic resolutions. We then report an enrichment of HOT regions near TAD boundaries and investigate the role of different transcription factors in determining domain borders at various resolutions. To further explore the interplay between domains organization and epigenomic features, we examine a distinctive pattern exhibited by the distribution of somatic mutations across boundaries. Overall, MrTADFinder provides a novel computational framework to explore the multi-scale structures stored in Hi-C contact maps.Author Summary The accommodation of the roughly 2m of DNA in the nuclei of mammalian cells results in an intricate structure, in which the topologically associating domains (TADs) formed by densely interacting genomic regions emerge as a fundamental structural unit. Identification of TADs is essential for understanding the role of 3D genome organization in gene regulation. By viewing the chromosomal contact map as a network, TADs correspond to the densely connected regions in the network. Motivated by this mapping, we propose a novel method, MrTADFinder, to identify TADs based on the concept of modularity in network science. Using MrTADFinder, we identify domains at various resolutions, and further explore the interplay between domains and other chromatin features like transcription factors binding and histone modifications at different resolutions. Overall, MrTADFinder provides a new computational framework to investigate the multiple length scales that are built inside the organization of the genome. ER -