HiCDOC: chromatin compartment prediction and differential analysis from Hi-C data with replicates
Year: 2026
Type: article
Abstract: Motivation : The spatial organization in the animal genome plays an essential role in regulating cellular functions, with A/B chromatin compartments reflecting broad differences in transcriptional and epigenetic activity. Hi-C enables genome-wide identification of such compartments, but robust differential analysis between groups of samples remains challenging. Existing approaches largely rely on Principal Component Analysis, which, applied on Hi-C matrices separately, requires heuristic sign choices to merge results and does not naturally incorporate replicates. Results : Here we present HiCDOC, a Bioconductor package for the prediction and differential analysis of chromatin compartments from Hi-C data with replicates. HiCDOC uses constrained k -means clustering to jointly analyze multiple Hi-C matrices, incorporating replicate information to enhance robustness, and provides empirical statistical support for predicted compartment switches. Applied to Hi-C datasets from human tissues and mouse cell lines, HiCDOC identified biologically relevant compartment changes supported by transcriptional differences. Comparisons with existing tools showed both overlap and complementarity, while a controlled benchmark with artificially introduced changes confirmed high sensitivity. Although extensively tested on pairwise comparisons, HiCDOC offers a flexible framework compatible with more complex designs and, in principle, with more than two compartment states. By combining replicate-aware clustering, automatic A/B assignment across chromosomes, extensive quality control, and statistical evaluation, HiCDOC provides an alternative and complementary approach to PCA-based methods for compartment analysis. HiCDOC thus expands the methodological toolkit for exploring 3D genome dynamics and its role in cellular processes. Availability : HiCDOC is implemented in R and C++, and is available on Bioconductor: https://bioconductor.org/packages/release/bioc/html/HiCDOC.html
Source: Peer Community Journal
Language: English
Subfield: Molecular Biology
Domain: Life Sciences
Open Access status: diamond
OpenAlex ID: W7234422412
