Hi-cGAN: Prediction of Hi-C interaction matrices with conditional generative adversarial networks
- 1Hannover Medical School
Abstract
Background: The three-dimensional organization of the genome is a fundamental aspect of its function and regulation. High-throughput chromosome conformation capture techniques, such as Hi-C, have revolutionized our understanding of spatial genome organization. However, 3C-based methods are resource-intensive and technically demanding. This has driven the development of computational approaches for predicting Hi-C interaction matrices. Hi-cGAN, a novel approach based on conditional generative adversarial networks, offers a computational alternative to extensive wet-lab work by predicting Hi-C interaction matrices. This computational approach contributes to a broader exploration and understanding of genome architecture. Findings: The network pairs a convolutional generator with a convolutional discriminator, evaluated across bin sizes, inputs and cell types. It predicts a whole genome as a cool file at bin sizes from 2 to 25 kb, where Akita, C.Origami and Epiphany emit fixed windows of 1 Mb, 2 Mb and 990 kb. With the input chosen on a validation chromosome, agreement approaches Epiphany's and stays below the sequence-based C.Origami and Akita: over Akita's 411 held-out windows the mean correlation is 0.238 against 0.506. Boundary and loop calls agree less closely, placing the maps at the domain scale. Conclusions: Chromatin factor occupancy determines a substantial part of contact structure, and two tracks capture most of it. The most informative track depends on the resolution: CTCF and the cohesin subunits at 5 to 10 kb, active histone marks at 25 kb. Transfer to an unseen cell type costs about 0.12 SCC, and which method leads depends on the measure.
Related
- Conditional Generative Adversarial Network for the simulation of seabird trajectories based on environmental data
- Generative Adversarial Networks for Omics Data: An Overview and Future Directions for Livestock Applications
- DHAFGan: A dense hybrid attention fusion generative adversarial network for infrared and visible image fusion
- Neural prediction decorrelation reveals that adversarial robustness substantially improves DNN prediction accuracy across the entire human auditory cortex
- Pep-PU-GAN: Positive-Unlabeled Adversarial Learning for Peptide Function Prediction
BibTeX
@misc{krauth2026hicgan,
title = {{Hi-cGAN: Prediction of Hi-C interaction matrices with conditional generative adversarial networks}},
author = {R. Krauth and A. Kumar and J. Wolff},
howpublished = {bioRxiv},
year = {2026},
month = aug,
doi = {10.64898/2026.08.30.748103},
url = {https://www.biorxiv.org/content/10.64898/2026.08.30.748103v1},
note = {CC BY-NC},
}
RIS
TY - UNPB
TI - Hi-cGAN: Prediction of Hi-C interaction matrices with conditional generative adversarial networks
AU - R. Krauth
AU - A. Kumar
AU - J. Wolff
PY - 2026
DA - 2026/08/30
PB - bioRxiv
DO - 10.64898/2026.08.30.748103
UR - https://www.biorxiv.org/content/10.64898/2026.08.30.748103v1
AB - Background: The three-dimensional organization of the genome is a fundamental aspect of its function and regulation. High-throughput chromosome conformation capture techniques, such as Hi-C, have revolutionized our understanding of spatial genome organization. However, 3C-based methods are resource-intensive and technically demanding. This has driven the development of computational approaches for predicting Hi-C interaction matrices. Hi-cGAN, a novel approach based on conditional generative adversarial networks, offers a computational alternative to extensive wet-lab work by predicting Hi-C interaction matrices. This computational approach contributes to a broader exploration and understanding of genome architecture. Findings: The network pairs a convolutional generator with a convolutional discriminator, evaluated across bin sizes, inputs and cell types. It predicts a whole genome as a cool file at bin sizes from 2 to 25 kb, where Akita, C.Origami and Epiphany emit fixed windows of 1 Mb, 2 Mb and 990 kb. With the input chosen on a validation chromosome, agreement approaches Epiphany's and stays below the sequence-based C.Origami and Akita: over Akita's 411 held-out windows the mean correlation is 0.238 against 0.506. Boundary and loop calls agree less closely, placing the maps at the domain scale. Conclusions: Chromatin factor occupancy determines a substantial part of contact structure, and two tracks capture most of it. The most informative track depends on the resolution: CTCF and the cohesin subunits at 5 to 10 kb, active histone marks at 25 kb. Transfer to an unseen cell type costs about 0.12 SCC, and which method leads depends on the measure.
ER -