Abstract
Three-dimensional genome organization controls cell-type-specific gene expression through chromatin interactions, yet systematic analysis across diverse cellular contexts remains limited by experimental constraints. Here we present Hi-Compass, a depth-aware deep learning framework that predicts cell-type-specific chromatin organization using only chromatin accessibility data as cell-type-specific input. By dynamically accommodating variability in sequencing depth, Hi-Compass enables robust predictions across the full spectrum of data scales, from sparse single-cell to high-coverage bulk profiles. Benchmarking shows that Hi-Compass achieves superior concordance with experimental Hi-C data compared to existing methods, with particularly strong recovery of high-confidence chromatin loops. Applied to peripheral blood and embryonic heart datasets, Hi-Compass resolves cell-type-specific chromatin interactions and systematically links disease-associated variants to putative target genes. The framework further enables spatially resolved chromatin interaction prediction in hippocampal tissue and demonstrates cross-species applicability through fine-tuning to mouse systems. Hi-Compass expands the capacity to study three-dimensional genome regulation across biological scales and species.
Data availability
The Hi-C, CTCF ChIP–seq and ATAC–seq datasets used in the study were all public data from the ENCODE, 4DN and GEO database, with the detailed information listed in the Supplementary Data 1-3. Cohesin HiChIP loop calls in GM12878 were obtained from Mumbach et al.43 and RAD21 ChIA-PET loops from Grubert et al.44 The single-cell ATAC-seq, Multiome and spatial ATAC-seq data were obtained from 10X Genomics (https://www.10xgenomics.com/datasets/), ENCODE and GEO database, with the detailed information listed in the Supplementary Data 4. GWAS SNPs data was obtained from GWAS Catalog70 (https://www.ebi.ac.uk/gwas/), eQTLs data from GTEx71 v10 (https://www.gtexportal.org/).
Code availability
The Hi-Compass framework was implemented in the ‘hicompass’ Python package, which is available at https://github.com/EndeavourSyc/Hi-Compass. The specific version of the code associated with this publication is archived in Zenodo (https://doi.org/10.5281/zenodo.19283061).
References
Bonev, B. & Cavalli, G. Organization and function of the 3D genome. Nat. Rev. Genet. 17, 661–678 (2016).
Dekker, J. & Mirny, L. The 3D genome as moderator of chromosomal communication. Cell 164, 1110–1121 (2016).
Kagey, M. H. et al. Mediator and cohesin connect gene expression and chromatin architecture. Nature 467, 430–435 (2010).
Lieberman-Aiden, E. et al. Comprehensive mapping of long-range interactions reveals folding principles of the human genome. science 326, 289–293 (2009).
Wang, H., Han, M. & Qi, L. S. Engineering 3D genome organization. Nat. Rev. Genet. 22, 343–360 (2021).
Krijger, P. H. L. & De Laat, W. Regulation of disease-associated gene expression in the 3D genome. Nat. Rev. Mol. cell Biol. 17, 771–782 (2016).
Jerkovic, I. & Cavalli, G. Understanding 3D genome organization by multidisciplinary methods. Nat. Rev. Mol. cell Biol. 22, 511–528 (2021).
Szabo, Q., Bantignies, F. & Cavalli, G. Principles of genome folding into topologically associating domains. Sci. Adv. 5, eaaw1668 (2019).
Fudenberg, G., Abdennur, N., Imakaev, M., Goloborodko, A. & Mirny, L.A. Emerging evidence of chromosome folding by loop extrusion. in Cold Spring Harbor symposia on quantitative biology Vol. 82 45-55 (Cold Spring Harbor Laboratory Press, 2017).
An, L. et al. OnTAD: hierarchical domain structure reveals the divergence of activity among TADs and boundaries. Genome Biol. 20, 1–16 (2019).
Li, A. et al. Decoding topologically associating domains with ultra-low resolution Hi-C data by graph structural entropy. Nat. Commun. 9, 3265 (2018).
da Costa-Nunes, J. A. & Noordermeer, D. TADs: Dynamic structures to create stable regulatory functions. Curr. Opin. Struct. Biol. 81, 102622 (2023).
Cavalheiro, G. R., Pollex, T. & Furlong, E. E. To loop or not to loop: what is the role of TADs in enhancer function and gene regulation? Curr. Opin. Genet. Dev. 67, 119–129 (2021).
Pollex, T. et al. Enhancer–promoter interactions become more instructive in the transition from cell-fate specification to tissue differentiation. Nat. Genet. 56, 686–696 (2024).
Szabo, Q. et al. Regulation of single-cell genome organization into TADs and chromatin nanodomains. Nat. Genet. 52, 1151–1157 (2020).
Zhang, S., Übelmesser, N., Barbieri, M. & Papantonis, A. Enhancer–promoter contact formation requires RNAPII and antagonizes loop extrusion. Nat. Genet. 55, 832–840 (2023).
Hsieh, T.-H. S. et al. Enhancer–promoter interactions and transcription are largely maintained upon acute loss of CTCF, cohesin, WAPL or YY1. Nat. Genet. 54, 1919–1932 (2022).
Rao, S. S. et al. A 3D map of the human genome at kilobase resolution reveals principles of chromatin looping. Cell 159, 1665–1680 (2014).
Zheng, H. & Xie, W. The role of 3D genome organization in development and cell differentiation. Nat. Rev. Mol. Cell Biol. 20, 535–550 (2019).
Zuin, J. et al. Nonlinear control of transcription through enhancer–promoter interactions. Nature 604, 571–577 (2022).
Chakraborty, C. et al. Rewiring of the promoter-enhancer interactome and regulatory landscape in glioblastoma orchestrates gene expression underlying neurogliomal synaptic communication. Nat. Commun. 14, 6446 (2023).
Haarhuis, J. H. et al. The cohesin release factor WAPL restricts chromatin loop extension. Cell 169, 693–707.e14 (2017).
Stik, G. et al. CTCF is dispensable for immune cell transdifferentiation but facilitates an acute inflammatory response. Nat. Genet. 52, 655–661 (2020).
Buenrostro, J. D., Giresi, P. G., Zaba, L. C., Chang, H. Y. & Greenleaf, W. J. Transposition of native chromatin for fast and sensitive epigenomic profiling of open chromatin, DNA-binding proteins and nucleosome position. Nat. methods 10, 1213–1218 (2013).
Corces, M. R. et al. An improved ATAC-seq protocol reduces background and enables interrogation of frozen tissues. Nat. methods 14, 959–962 (2017).
Cusanovich, D. A. et al. The cis-regulatory dynamics of embryonic development at single-cell resolution. Nature 555, 538–542 (2018).
Tang, Z. et al. CTCF-mediated human 3D genome architecture reveals chromatin topology for transcription. Cell 163, 1611–1627 (2015).
Nora, E. P. et al. Targeted degradation of CTCF decouples local insulation of chromosome domains from genomic compartmentalization. Cell 169, 930–944 (2017).
Fudenberg, G. et al. Formation of chromosomal domains by loop extrusion. Cell Rep. 15, 2038–2049 (2016).
Phillips-Cremins, J. E. et al. Architectural protein subclasses shape 3D organization of genomes during lineage commitment. Cell 153, 1281–1295 (2013).
Liu, S. et al. Cell type–specific 3D-genome organization and transcription regulation in the brain. Sci. Adv. 11, eadv2067 (2025).
Wang, R. et al. MyoD is a 3D genome structure organizer for muscle cell identity. Nat. Commun. 13, 205 (2022).
Tan, J. et al. Cell-type-specific prediction of 3D chromatin organization enables high-throughput in silico genetic screening. Nat. Biotechnol. 41, 1140–1150 (2023).
Nagano, T. et al. Single-cell Hi-C reveals cell-to-cell variability in chromosome structure. Nature 502, 59–64 (2013).
Schwessinger, R. et al. DeepC: predicting 3D genome folding using megabase-scale transfer learning. Nat. methods 17, 1118–1124 (2020).
Fudenberg, G., Kelley, D. R. & Pollard, K. S. Predicting 3D genome folding from DNA sequence with Akita. Nat. methods 17, 1111–1117 (2020).
Avsec, Ž et al. AlphaGenome: advancing regulatory variant effect prediction with a unified DNA sequence model. bioRxiv 06, 661532 (2025).
Yang, R. et al. Epiphany: predicting Hi-C contact maps from 1D epigenomic signals. Genome Biol. 24, 134 (2023).
Gao, V. R. et al. ChromaFold predicts the 3D contact map from single-cell chromatin accessibility. Nat. Commun. 15, 9432 (2024).
Gentner, D. & Markman, A. B. Defining structural similarity. J. Cogn. Sci. 6, 1–20 (2006).
Wang, Z. & Li, Q. Information content weighting for perceptual image quality assessment. IEEE Trans. image Process. 20, 1185–1198 (2010).
Crane, E. et al. Condensin-driven remodelling of X chromosome topology during dosage compensation. Nature 523, 240–244 (2015).
Mumbach, M. R. et al. HiChIP: efficient and sensitive analysis of protein-directed genome architecture. Nat. methods 13, 919–922 (2016).
Grubert, F. et al. Landscape of cohesin-mediated chromatin loops in the human genome. Nature 583, 737–743 (2020).
Peripheral blood mononuclear cells (PBMCs) from a healthy donor. Epi ATAC dataset analyzed using Cell Ranger ATAC 1.2.0. https://www.10xgenomics.com/datasets/10-k-peripheral-blood-mononuclear-cells-pbm-cs-from-a-healthy-donor-1-standard-1-2-0. 10X Genomics (2020).
Giorgiutti, S., Rottura, J., Korganow, A.-S. & Gies, V. CXCR4: from B-cell development to B cell–mediated diseases. Life Sci. Alliance 7, e202302465 (2024).
Nie, Y. et al. The Role of CXCR4 in maintaining peripheral B cell compartments and humoral immunity. J. Exp. Med. 200, 1145 (2004).
Uckun, F. M. Regulation of human B-cell ontogeny. Blood 76, 1908–1923 (1990).
Mogilenko, D. A. et al. Comprehensive profiling of an aging immune system reveals clonal GZMK+ CD8+ T cells as conserved hallmark of inflammaging. Immunity 54, 99–115 (2021).
Zhang, R. et al. Association between CDK8 gene polymorphisms and dilated cardiomyopathy in a Chinese Han population. Cardiovasc. Diag. Ther. 11, 1036 (2021).
Hall, D. D. et al. Ectopic expression of Cdk8 induces eccentric hypertrophy and heart failure. JCI insight 2, e92476 (2017).
Spielmann, N. et al. Extensive identification of genes involved in congenital and structural heart disorders and cardiomyopathy. Nat. Cardiovasc. Res. 1, 157–173 (2022).
Funakoshi, S. et al. Generation of mature compact ventricular cardiomyocytes from human pluripotent stem cells. Nat. Commun. 12, 3155 (2021).
Hocker, J. D. et al. Cardiac cell type–specific gene regulatory programs and disease risk association. Sci. Adv. 7, eabf1444 (2021).
Lewis, S. M. et al. Spatial omics and multiplexed imaging to explore cancer biology. Nat. methods 18, 997–1012 (2021).
Bressan, D., Battistoni, G. & Hannon, G. J. The dawn of spatial omics. Science 381, eabq4964 (2023).
Vandereyken, K., Sifrim, A., Thienpont, B. & Voet, T. Methods and applications for single-cell and spatial multi-omics. Nat. Rev. Genet. 24, 494–515 (2023).
Moses, L. & Pachter, L. Museum of spatial transcriptomics. Nat. Methods 19, 534–546 (2022).
Tian, L., Chen, F. & Macosko, E. Z. The expanding vistas of spatial transcriptomics. Nat. Biotechnol. 41, 773–782 (2023).
Li, H. et al. Spatially resolved genome-wide joint profiling of epigenome and transcriptome with spatial-ATAC-RNA-seq and spatial-CUT&Tag-RNA-seq. Nat. Protoc. 20, 2383–2417 (2025).
Zhang, D. et al. Spatial epigenome–transcriptome co-profiling of mammalian tissues. Nature 616, 113–122 (2023).
Davis, C. A. et al. The encyclopedia of DNA elements (ENCODE): data portal update. Nucleic acids Res. 46, D794–D801 (2018).
Reiff, S. B. et al. The 4D nucleome data portal as a resource for searching and visualizing curated nucleomics data. Nat. Commun. 13, 2365 (2022).
Barrett, T. et al. NCBI GEO: archive for functional genomics data sets—update. Nucleic Acids Res. 41, D991–D995 (2012).
Li, H. & Durbin, R. Fast and accurate short read alignment with Burrows–Wheeler transform. Bioinformatics 25, 1754–1760 (2009).
Abdennur, N., Fudenberg, G., Flyamer, I. M., Galitsyna, A. A., Goloborodko, A., Imakaev, M. & Venev, S. V. Pairtools: from sequencing data to chromosome contacts. PLoS Comput. Biol. 20, e1012164 (2024).
Abdennur, N. & Mirny, L. A. Cooler: scalable storage for Hi-C data and other genomically labeled arrays. Bioinformatics 36, 311–316 (2020).
Quinlan, A. R. & Hall, I. M. BEDTools: a flexible suite of utilities for comparing genomic features. Bioinformatics 26, 841–842 (2010).
Perez, G. et al. The UCSC Genome Browser database: 2025 update. Nucleic Acids Res. 53, D1243–D1249 (2025).
Buniello, A. et al. The NHGRI-EBI GWAS catalog of published genome-wide association studies, targeted arrays and summary statistics 2019. Nucleic Acids Res. 47, D1005–D1012 (2019).
Consortium, G. The GTEx Consortium atlas of genetic regulatory effects across human tissues. Science 369, 1318–1330 (2020).
Acknowledgements
We gratefully acknowledge the High-performance Computing Platform of Peking University for conducting the data analyses. The deep learning model training was carried out on Tianhe new generation supercomputer at National Supercomputer Center in Tianjin. Funding This work was supported by the National Natural Science Foundation of China (32270683 and 32470662); the Beijing Natural Science Foundation (5242006); the Fundamental Research Funds for the Central Universities (BMU2021YJ064) to H.J.W.; CAMS Innovation Fund for Medical Sciences (2021-I2M-5-003) to M.X.; the Science Foundation of Peking University Cancer Hospital (ZY202418) to Y.C.S.; the China Postdoctoral Science Foundation (2024M750125) to N.N.W.
Author information
Authors and Affiliations
Contributions
Y.C.S., W.J.J., K.W.C., and H.J.W. conceived and designed the study. Y.C.S. developed the model and performed the main computational analyses. W.J.J. and K.W.C. contributed to data processing and algorithm development. N.N.W., F.T.L., H.J.W., and R.X.G. contributed to benchmarking, data curation, validation and biological interpretation. Z.Y.K., J.L.Z., A.L., and H.W.Z. assisted with data collection. Y.C.S. and H.J.W. wrote the manuscript. Y.J.W., M.X., and H.J.W. supervised the study. All authors reviewed and approved the final manuscript.
Corresponding authors
Ethics declarations
Competing interests
The authors declare no competing interests.
Peer review
Peer review information
Nature Communications thanks Yaqiang Cao and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. A peer review file is available.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary information
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Sun, YC., Jiang, WJ., Cai, KW. et al. Hi-Compass: a depth-aware deep learning framework for predicting cell-type-specific 3D genome organization from single-cell to spatial resolution. Nat Commun (2026). https://doi.org/10.1038/s41467-026-71877-z
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41467-026-71877-z
