DNABERT2-CAMP: A Hybrid Transformer-CNN Model for E. coli Promoter Recognition
Genes · 2025
DOI: 10.3390/genes17010027 ↗Kurz gefasst
Fachartikel in der Zeitschrift Genes, 2025. Verfasst von Hua-Lin Xu und 3 weiteren Autor:innen. Beteiligte Einrichtung: Department of Intelligent Technology, Tianjin Polytechnic University, Tianjin 300340, China. Hauptthemen laut MEDLINE: Promoter Regions, Genetic, Escherichia coli, Computational Biology. Die Arbeit nennt 37 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Special Project of Natural Science Foundation of China, Special Project of Scientific and Technological Basic Resources Survey of the Ministry of Science and Technology of China.
Abstract (Originalsprache)
Background
Accurate recognition of promoter sequences in Escherichia coli is fundamental for understanding gene regulation and engineering synthetic biological systems. However, existing computational methods struggle to simultaneously model long-range genomic dependencies and fine-grained local motifs, particularly the degenerate −10 and −35 elements of σ70 promoters. To address this gap, we propose DNABERT2-CAMP, a novel hybrid deep learning framework designed to integrate global contextual understanding with high-resolution local motif detection for robust promoter identification.
Methods
We constructed a balanced dataset of 8720 experimentally validated and negative 81-bp sequences from RegulonDB, literature, and the E. coli K-12 genome. Our model combines a pre-trained DNABERT-2 Transformer for global sequence encoding with a custom CAMP module (CNN-Attention-Mean Pooling) for local feature refinement. We evaluated performance using 5-fold cross-validation and an independent external test set, reporting standard metrics including accuracy, ROC AUC, and Matthews correlation coefficient (MCC).
Results
DNABERT2-CAMP achieved 93.10% accuracy and 97.28% ROC AUC in cross-validation, outperforming existing methods including DNABERT. On an independent test set, it maintained strong generalization (89.83% accuracy, 92.79% ROC AUC). Interpretability analyses confirmed biologically plausible attention over canonical promoter regions and CNN-identified AT-rich/-35-like motifs.
Conclusions
DNABERT2-CAMP demonstrates that synergistically combining pre-trained Transformers with convolutional motif detection significantly improves promoter recognition accuracy and interpretability. This framework offers a powerful, generalizable tool for genomic annotation and synthetic biology applications.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Intelligent Technology, Tianjin Polytechnic University, Tianjin 300340, China.
- School of Artificial Intelligence, Tianjin University, No. 135 Yaguan Road, Haihe Education Park, Tianjin 300354, China.
- Key Laboratory of Systems Bioengineering (Ministry of Education), No. 135 Yaguan Road, Haihe Education Park, Tianjin 300354, China.
- School of Computer Science and Technology, Tianjin University, No. 135 Yaguan Road, Haihe Education Park, Tianjin 300354, China.
ORCID-Profile: Xiu-Jun Gong
Förderung
- Special Project of Natural Science Foundation of China
- Special Project of Scientific and Technological Basic Resources Survey of the Ministry of Science and Technology of China
Zitieren
APA
Xu, H. L., Gong, X. J., Yu, H., & Wang, Y. K. (2025). DNABERT2-CAMP: A Hybrid Transformer-CNN Model for E. coli Promoter Recognition. Genes. https://doi.org/10.3390/genes17010027
Vancouver
Xu HL, Gong XJ, Yu H, Wang YK. DNABERT2-CAMP: A Hybrid Transformer-CNN Model for E. coli Promoter Recognition. Genes. 2025. doi:10.3390/genes17010027
BibTeX
@article{Xu2025,
author = {Xu, Hua-Lin and Gong, Xiu-Jun and Yu, Hua and Wang, Ying-Kai},
title = {DNABERT2-CAMP: A Hybrid Transformer-CNN Model for E. coli Promoter Recognition},
journal = {Genes},
year = {2025},
doi = {10.3390/genes17010027}
}
EUVIMED