Chiron3D: an interpretable deep learning framework for understanding the DNA code of chromatin looping
Bioinformatics · 2026
DOI: 10.1093/bioinformatics/btag492 ↗Kurz gefasst
Fachartikel in der Zeitschrift Bioinformatics, 2026. Verfasst von Sebastian Hönig und 4 weiteren Autor:innen. Beteiligte Einrichtung: Department of Computer Science, ETH Zurich, Zurich, Switzerland. Hauptthemen laut MEDLINE: Chromatin, Deep Learning, DNA, Computational Biology. Die Arbeit nennt 25 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Swiss Government Excellence Scholarship, Swiss Data Science Center, collaborative projects grant.
Abstract (Originalsprache)
Motivation Three-dimensional folding of the genome into structures such as chromatin loops is essential for gene regulation. Current experimental methods for mapping these structures, like Hi-C and HiChIP, are labor-intensive and require repeated assays to test hypothesized mutation effects. This motivates the need for predictive approaches that reveal the sequence determinants of chromatin loops. Results In this work, we present a novel and interpretable computational pipeline for predicting CTCF-mediated chromatin loops. We propose Chiron3D, a DNA-only model trained in a cell-type specific manner to predict CTCF HiChIP contact maps. By leveraging pre-trained embeddings from a foundation model, our approach is competitive with baselines that take CTCF ChIP-seq as additional input, while enabling nucleotide-level attribution to the input DNA sequence. Using our framework, we provide likely mechanistic insights into the physical control of loop dynamics. Specifically, we find that the strength of the loop extrusion anchorage site is largely governed by the amount and binding affinity of CTCF sites at the boundaries. Furthermore, we reveal that loop stability is regulated by the amount of intra-loop CTCF binding sites, where fewer intra-loop sites are associated with greater loop stability. Using targeted, single-nucleotide edit simulations with Chiron3D, we show that both loop strength and stability can be precisely controlled. Together, these results provide novel mechanistic insights into the physical control of genome organization and highlight the potential of decoding the DNA sequence logic in silico. Availability The Chiron3D pipeline is made available at https://github.com/BoevaLab/Chiron3D.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Substanzen
Autor:innen und Institutionen
- Department of Computer Science, ETH Zurich, Zurich, Switzerland.
- Swiss Institute of Bioinformatics, Lausanne, Switzerland.
- Balgrist University Hospital, Faculty of Medicine, University of Zurich, Zurich, Switzerland.
- Université Paris Cité, Institut Cochin, INSERM U1016, Paris, France.
ORCID-Profile: Aayush Grover, Didier Surdez, Valentina Boeva
Förderung
- Swiss Government Excellence Scholarship
- Swiss Data Science Center
- collaborative projects grant
- Swiss National Science Foundation
- Swiss Cancer League
- Mercier Foundation
- Balgrist Foundation
- Resortho Foundation
Zitieren
APA
Hönig, S., Grover, A., Neri, P., Surdez, D., & Boeva, V. (2026). Chiron3D: an interpretable deep learning framework for understanding the DNA code of chromatin looping. Bioinformatics. https://doi.org/10.1093/bioinformatics/btag492
Vancouver
Hönig S, Grover A, Neri P, Surdez D, Boeva V. Chiron3D: an interpretable deep learning framework for understanding the DNA code of chromatin looping. Bioinformatics. 2026. doi:10.1093/bioinformatics/btag492
BibTeX
@article{Hnig2026,
author = {Hönig, Sebastian and Grover, Aayush and Neri, Piero and Surdez, Didier and Boeva, Valentina},
title = {Chiron3D: an interpretable deep learning framework for understanding the DNA code of chromatin looping},
journal = {Bioinformatics},
year = {2026},
doi = {10.1093/bioinformatics/btag492}
}
EUVIMED