GrapHiCΒΆ

GrapHiC is a python library for constructing graph representations of Hi-C data and integrating such representations with multi-omics data for computational analysis. The library interfaces with the popular Cooler Hi-C format (although we plan to add functionality for other Hi-C formats in future) and Pytorch Geometric by default. As feature engineering is a vital step in a machine learning project, the library is designed to be highly flexible, allowing the user to parameterise the graph construction, creating lists of subgraphs from sets of genomic regions or by specifying particular genomic loci. GrapHiC is also designed to facilitate network-based and graph-theoretic analyses of protein structures in a high-throughput manner.