EAGP: an efficient generative augmentation framework for phage protein classification under severe class imbalance
Bioinformatics · 01.06.2026 · 42 (6)
DOI: 10.1093/bioinformatics/btag373 ↗Kurz gefasst
Fachartikel in der Zeitschrift Bioinformatics, online erschienen am 15.06.2026. Verfasst von Jiaru Li und 4 weiteren Autor:innen. Beteiligte Einrichtung: Faculty of Information Science and Engineering, Ocean University of China, Qingdao, Shandong 266100, China. Hauptthemen laut MEDLINE: Bacteriophages, Viral Proteins, Computational Biology, Software. Die Arbeit nennt 30 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China.
Abstract (Originalsprache)
Motivation The accurate classification of phage proteins is critical for advancing bacteriophage research. Despite the proliferation of machine learning approaches in this domain, the persistent issue of data imbalance continues to hinder performance, particularly for rare protein sequences. Previous attempts to address this by re-weighting minority classes have faced limitations due to insufficient feature extraction capabilities. Results In this paper, we introduce EAGP, a novel approach that integrates a generative model—functionally equivalent to a WGAN yet tailored for one-dimensional data—with the Evolutionary Scale Modeling (ESM) protein large language model for robust feature extraction. EAGP exhibits exceptional performance in binary classification and protein function annotation tasks. Crucially, our method not only improves overall classification efficacy but also significantly alleviates the performance degradation typically observed in minority classes. Availability and Implementation The data and code underlying this article are available in GitHub at https://github.com/Innerly/EAGP and have been archived on Zenodo at https://doi.org/10.5281/zenodo.19928069.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Substanzen
Autor:innen und Institutionen
- Faculty of Information Science and Engineering, Ocean University of China, Qingdao, Shandong 266100, China.
- Institute of Fundamental and Frontier Sciences, University of Electronic Science and Technology of China, Chengdu, 610054, China.
- Institute of Digital Health, Yangtze Delta Region Institute (Quzhou), University of Electronic Science and Technology of China, Quzhou 324003, China.
- Shanghai Sixth People's Hospital Affiliated to Shanghai Jiao Tong University School of Medicine, Shanghai, 200233, China.
ORCID-Profile: Jiaru Li, Yansu Wang, Quan Zou
Förderung
- National Natural Science Foundation of China
Zitieren
APA
Li, J., Li, H., Wang, Y., Zou, Q., & Zhu, H. (2026). EAGP: an efficient generative augmentation framework for phage protein classification under severe class imbalance. Bioinformatics, 42(6). https://doi.org/10.1093/bioinformatics/btag373
Vancouver
Li J, Li H, Wang Y, Zou Q, Zhu H. EAGP: an efficient generative augmentation framework for phage protein classification under severe class imbalance. Bioinformatics. 2026;42(6). doi:10.1093/bioinformatics/btag373
BibTeX
@article{Li2026,
author = {Li, Jiaru and Li, Haoxiang and Wang, Yansu and Zou, Quan and Zhu, Hongling},
title = {EAGP: an efficient generative augmentation framework for phage protein classification under severe class imbalance},
journal = {Bioinformatics},
year = {2026},
volume = {42},
number = {6},
doi = {10.1093/bioinformatics/btag373}
}
EUVIMED