Computer vision for pattern detection in chromosome contact maps

Abstract

Abstract Chromosomes of all species studied so far display a variety of higher order organizational features such as domains or loops often associated to biological functions and visible on Hi-C contact maps. We developed Chromosight , an algorithm inspired from computer vision that can detect patterns in Hi-C maps. Chromosight has greater sensitivity than existing methods, while being faster and applicable to any type of genomes, including bacteria, viruses, yeasts and mammals. Code and documentation: https://github.com/koszullab/chromosight