Abstract

Chromatin organization in three-dimension (3D) is emerging as key to understanding crucial biological functions in heath and disease. While high-throughput 3D genome technologies have been rapidly developed in recent years, few methods exist to effectively analyze the noisy and highly structured chromatin-chromatin interaction data. I will introduce new statistical methods for estimating 3D chromatin structures and detecting chromosomal abnormalities from genome-wide interaction data. Through simulations and real data analyses, we demonstrate improved accuracy and power of our methods over existing approaches.