Integrative Modeling of Multi-Track Hi-C Data for Genome-Scale Reconstruction of 3D Chromatin Structure

• Activity Number: 195
• Type: Contributed
• Date/Time: Monday, August 10, 2015 : 10:30 AM to 12:20 PM
• Sponsor: International Chinese Statistical Association
• Author(s): Chenchen Zou* and Yuping Zhang and Zhengqing Ouyang
• Companies: The Jackson Laboratory for Genomic Medicine and University of Connecticut and The Jackson Laboratory for Genomic Medicine
• Keywords: Hi-C ; genome-wide ; chromatin structure ; integrative modeling ; generalized linear model

Abstract

Genome-wide chromatin conformation-capturing (Hi-C) has been widely used to study 3D interactions and structures of the genome. However, few computational approaches are existing to quantitatively analyze Hi-C data, thus hindering the investigation of the association between 3D chromatin structure and genome function. Here, we present a flexible tool that can jointly analyze multiple contact maps from different Hi-C experiments to infer the 3D chromatin structures of the genome. We use a generalized linear model framework to globally search for the latent structure underlying the cleavage footprints of different restriction enzymes. The proposed model is robust, accurate, and outperforms existing computational tools when evaluated on independent experimental datasets for a number of diverse cell types. The reconstructed 3D chromatin structures provide new insights on the roles of the chromatin remodeling protein CTCF.