HCMB

The Hi-C Matrix Balancing (HCMB) is a software tool to process Hi-C (high-throughput genome-wide chromosome conformation capture) data: the normalization of chromatin contact matrices. Hi-C is a pivotal method for studying chromosomal interactions, enabling the extraction of valuable biological information such as the P(s) curve, Topologically Associating Domains (TADs), A/B compartments, and other significant biological signals. The normalization process is essential to eliminate systematic and technical biases in the data stemming from mappability variations, GC content, and restriction fragment lengths. A particular issue with Hi-C data is its high sparsity, which complicates the correction process and underscores the need for a stable and efficient normalization method.

HCMB introduces an innovative algorithm that combines an iterative solution of equations with a linear search and projection strategy, specifically designed to tackle the issue of high sparsity in Hi-C data normalization. This method has been tested on both simulated and experimental data, demonstrating its robustness and efficiency in normalizing Hi-C data. It successfully preserves the biologically relevant features of Hi-C data, even in cases of extreme sparsity.