M3Site: multiclass multimodal learning for protein active site identification and classification
Briefings in Bioinformatics · 2025
DOI: 10.1093/bib/bbaf590 ↗Kurz gefasst
Fachartikel in der Zeitschrift Briefings in Bioinformatics, 2025. Verfasst von Song Ouyang und 9 weiteren Autor:innen. Beteiligte Einrichtung: Renmin Hospital of Wuhan University, Zhang Road and Jiefang Road, Wuhan, Hubei 430060, China. Hauptthemen laut MEDLINE: Proteins, Catalytic Domain, Computational Biology, Software. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 44 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Interdisciplinary Innovative Talents Foundation from Renmin Hospital of Wuhan University, Fundamental Research Funds for the Central Universities, National Natural Science Foundation of China.
Abstract (Originalsprache)
Accurately identifying and classifying protein active sites is crucial for understanding protein mechanisms, drug design, and synthetic biology. Current methods often rely on binary classification and single-modal data, limiting their scope. To address these limitations, we propose M$^{3}$Site, a multimodal framework that integrates protein sequence embeddings, structural graph representations, and functional text annotations for residue-level, multiclass active site prediction. Built upon a curated dataset of 25 883 proteins sourced from UniProt and AlphaFold2, M$^{3}$Site leverages pretrained protein language models, equivariant graph neural networks, and biomedical language models for feature extraction. The function informed cross-attention module enables cross-modal feature fusion, while the adaptive weighted fusion mechanism balances modality contributions. A compound loss function tackles class imbalance, ensuring robust performance. Experimental results show M$^{3}$Site significantly outperforms existing models, and an interactive application has been developed to enhance its practical utility for predictions and visualizations. The dataset, source code for experiments, and interactive application are publicly available at https://github.com/Gift-OYS/M3Site.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Substanzen
Autor:innen und Institutionen
- Renmin Hospital of Wuhan University, Zhang Road and Jiefang Road, Wuhan, Hubei 430060, China.
- School of Computer Science, National Engineering Research Center for Multimedia Software, Wuhan University, Bayi Road, Wuhan, Hubei 430072, China.
- BioGeometry, North Haidian 2nd Street, Beijing 100083, China.
- Mila - Québec AI Institute, Rue Saint-Urbain, Montréal, Québec H2S 3H1, Canada.
- Department of Computer Science and Operations Research, Université de Montréal, Bd Édouard-Montpetit, Montréal, Québec H3T 1J4, Canada.
- Peking Union Medical College Hospital, No. 1 Shuaifuyuan Wangfujing Dongcheng District, Beijing 100730, China.
ORCID-Profile: Song Ouyang, Yong Luo
Förderung
- Interdisciplinary Innovative Talents Foundation from Renmin Hospital of Wuhan University
- Fundamental Research Funds for the Central Universities
- National Natural Science Foundation of China
- Innovative Research Group Project of Hubei Province
- Science and Technology Major Project of Hubei Province
Zitieren
APA
Ouyang, S., Luo, Y., Cai, H., Su, K., Liao, F., Zhan, N., Zhao, H., Yin, T., Zhao, L., & Shan, D. (2025). M3Site: multiclass multimodal learning for protein active site identification and classification. Briefings in Bioinformatics. https://doi.org/10.1093/bib/bbaf590
Vancouver
Ouyang S, Luo Y, Cai H, Su K, Liao F, Zhan N, et al. M3Site: multiclass multimodal learning for protein active site identification and classification. Briefings in Bioinformatics. 2025. doi:10.1093/bib/bbaf590
BibTeX
@article{Ouyang2025,
author = {Ouyang, Song and Luo, Yong and Cai, Huiyu and Su, Kehua and Liao, Fei and Zhan, Na and Zhao, Huangxuan and Yin, Tailang and Zhao, Lin and Shan, Dongjing},
title = {M3Site: multiclass multimodal learning for protein active site identification and classification},
journal = {Briefings in Bioinformatics},
year = {2025},
doi = {10.1093/bib/bbaf590}
}
EUVIMED