UniversalEPI: robust prediction of cell type-specific and differential chromatin interactions from DNA sequence and chromatin accessibility
Nucleic Acids Research · 2026
DOI: 10.1093/nar/gkag485 ↗Kurz gefasst
Fachartikel in der Zeitschrift Nucleic Acids Research, 2026. Verfasst von Aayush Grover und 12 weiteren Autor:innen. Beteiligte Einrichtung: Department of Computer Science, ETH Zurich, Zurich 8092, Switzerland. Hauptthemen laut MEDLINE: Chromatin, Enhancer Elements, Genetic, Computational Biology. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 85 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0). Gefördert von: Swiss Data Science Center, Swiss Government Excellence Scholarship, ETH Zurich.
Abstract (Originalsprache)
Enhancer–promoter interactions (EPIs) play a central role in gene regulation, but experimental techniques such as Hi-C for mapping these interactions remain costly and labor-intensive. Computational methods have been developed to predict EPIs in silico from DNA sequence and chromatin information; however, there are major challenges with the generalizability and accuracy of predictions by existing methods across cell types and conditions unseen during model training. We developed and validated UniversalEPI, an attention-based deep ensemble model that predicts EPIs up to 2 Mb apart using only DNA sequence and chromatin accessibility (ATAC-seq) data. Unlike models that reconstruct full Hi-C contact maps, UniversalEPI focuses on biologically relevant, sparse chromatin interactions between accessible regulatory elements. It generalizes across both bulk and single-cell ATAC-seq-derived pseudo-bulk datasets, delivering state-of-the-art performance while using fewer input modalities than existing approaches. By modeling predictive uncertainty, UniversalEPI enables statistically robust differential analysis of chromatin interactions across conditions. We demonstrate its utility by tracking dynamic EPIs during human macrophage activation and identifying regulatory differences between cancer cell states in esophageal adenocarcinoma. By providing precalculated Hi-C predictions for 157 ENCODE datasets, UniversalEPI expands the scope and applicability of in silico 3D genome modeling for studying gene regulation in development and disease.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Substanzen
Autor:innen und Institutionen
- Department of Computer Science, ETH Zurich, Zurich 8092, Switzerland.
- Swiss Institute of Bioinformatics, Lausanne 1015, Switzerland.
- Swiss Data Science Center, EPF Lausanne and ETH Zurich, Zurich 8092, Switzerland.
- Department of Medical Oncology, Dana-Farber Cancer Institute, Boston, MA 02215, United States.
- Cancer Program, Broad Institute of MIT and Harvard, Cambridge, MA 02142, United States.
- Division of Medical Sciences, Harvard University, Boston, MA 02115, United States.
ORCID-Profile: Eliezer M Van Allen, Valentina Boeva
Förderung
- Swiss Data Science Center
- Swiss Government Excellence Scholarship
- ETH Zurich
Zitieren
APA
Grover, A., Zhang, L., Muser, T., Häfliger, S., Wang, M., Yates, J., Indilewitsch, M. C., Wang, Y., Van Allen, E. M., Theis, F. J., Ibarra, I. L., Krymova, E., & Boeva, V. (2026). UniversalEPI: robust prediction of cell type-specific and differential chromatin interactions from DNA sequence and chromatin accessibility. Nucleic Acids Research. https://doi.org/10.1093/nar/gkag485
Vancouver
Grover A, Zhang L, Muser T, Häfliger S, Wang M, Yates J, et al. UniversalEPI: robust prediction of cell type-specific and differential chromatin interactions from DNA sequence and chromatin accessibility. Nucleic Acids Research. 2026. doi:10.1093/nar/gkag485
BibTeX
@article{Grover2026,
author = {Grover, Aayush and Zhang, Lin and Muser, Till and Häfliger, Simeon and Wang, Minjia and Yates, Josephine and Indilewitsch, Marie-Claire and Wang, Yizhen and Van Allen, Eliezer M and Theis, Fabian J and Ibarra, Ignacio L and Krymova, Ekaterina and Boeva, Valentina},
title = {UniversalEPI: robust prediction of cell type-specific and differential chromatin interactions from DNA sequence and chromatin accessibility},
journal = {Nucleic Acids Research},
year = {2026},
doi = {10.1093/nar/gkag485}
}
EUVIMED