EPINTLM: enhancer–promoter prediction with pretrained k-mer embeddings and residual cross-attention
Briefings in Bioinformatics · 2026
DOI: 10.1093/bib/bbag064 ↗Kurz gefasst
Fachartikel in der Zeitschrift Briefings in Bioinformatics, 2026. Verfasst von Thi Lan Nguyen und 5 weiteren Autor:innen. Beteiligte Einrichtung: VNU University of Engineering and Technology, Vietnam National University, 144 Xuan Thuy, Cau Giay, Hanoi 100000, Vietnam. Hauptthemen laut MEDLINE: Enhancer Elements, Genetic, Promoter Regions, Genetic, Deep Learning, Computational Biology. Die Arbeit nennt 22 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Science and Technology Council, Taiwan.
Abstract (Originalsprache)
Enhancer–promoter interactions (EPIs) play an important role in gene regulation, yet experimental mapping remains costly and limited in coverage. As a result, computational approaches are commonly evaluated under curated benchmark datasets, which pose challenges related to long-range sequence modeling, multimodal feature integration, and reproducible preprocessing. In this study, we present EPINTLM (Enhancer–Promoter Interaction Nucleotide Transformer Large Model), a deep learning framework designed to investigate architectural strategies for EPI prediction under standardized benchmark settings. EPINTLM integrates DNA sequence representations and genomic features by leveraging pretrained k-mer embeddings from the Nucleotide Transformer and explicitly modeling intra- and inter-sequence dependencies through residual self-attention and bidirectional cross-attention. We additionally introduce a unified preprocessing pipeline to improve training efficiency and reproducibility, and perform post hoc motif analysis to provide limited interpretability of learned sequence patterns. Evaluated on a widely used benchmark across six human cell lines, EPINTLM achieves competitive area under the receiver operating characteristic curve (AUROC) and area under the precision-recall curve (AUPR) performance relative to existing methods, with ablation studies highlighting the contributions of cross-attention and residual aggregation. These results demonstrate the utility of explicit cross-attention designs for paired regulatory sequence modeling within current benchmark constraints.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- VNU University of Engineering and Technology, Vietnam National University, 144 Xuan Thuy, Cau Giay, Hanoi 100000, Vietnam.
- AIBioMed Research Group, Taipei Medical University, 250 Wuxing Street, Taipei 110, Taiwan.
- International Master/Ph.D. Program in Medicine, College of Medicine, Taipei Medical University, 250 Wuxing Street, Taipei 110, Taiwan.
- In-Service Master Program in Artificial Intelligence in Medicine, College of Medicine, Taipei Medical University, 250 Wuxing Street, Taipei 110, Taiwan.
- Translational Imaging Research Center, Taipei Medical University Hospital, 252 Wuxing Street, Taipei 110, Taiwan.
ORCID-Profile: Nguyen Quoc Khanh Le
Förderung
- National Science and Technology Council, Taiwan
Zitieren
APA
Nguyen, T. L., Kha, H. Q., Nguyen, P. K., Le, M. H. N., Le, D. T., & Quoc Khanh Le, N. (2026). EPINTLM: enhancer–promoter prediction with pretrained k-mer embeddings and residual cross-attention. Briefings in Bioinformatics. https://doi.org/10.1093/bib/bbag064
Vancouver
Nguyen TL, Kha HQ, Nguyen PK, Le MHN, Le DT, Quoc Khanh Le N. EPINTLM: enhancer–promoter prediction with pretrained k-mer embeddings and residual cross-attention. Briefings in Bioinformatics. 2026. doi:10.1093/bib/bbag064
BibTeX
@article{Nguyen2026,
author = {Nguyen, Thi Lan and Kha, Hien Quang and Nguyen, Phat Ky and Le, Minh Huu Nhat and Le, Duc-Trong and Quoc Khanh Le, Nguyen},
title = {EPINTLM: enhancer–promoter prediction with pretrained k-mer embeddings and residual cross-attention},
journal = {Briefings in Bioinformatics},
year = {2026},
doi = {10.1093/bib/bbag064}
}
EUVIMED