Dual-channel graph learning reveals similarity and complementarity in protein-protein interaction networks
PLOS Computational Biology · 20.05.2026 · 22 (5) · e1013941
DOI: 10.1371/journal.pcbi.1013941 ↗Kurz gefasst
Fachartikel in der Zeitschrift PLOS Computational Biology, online erschienen am 20.05.2026. Verfasst von Tao Tang und 7 weiteren Autor:innen. Beteiligte Einrichtung: School of Modern Posts, Nanjing University of Posts and Telecommunications, Jiangsu, China. Hauptthemen laut MEDLINE: Protein Interaction Mapping, Computational Biology, Protein Interaction Maps. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 52 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Nature Science Foundation of China.
Abstract (Originalsprache)
Protein-protein interactions (PPIs) are governed by two fundamental interfacial mechanisms: similarity-driven, often involving symmetric structural motifs, and complementarity-driven, arising from geometric and physicochemical matching between binding surfaces. Despite their biological significance, computational models have largely overlooked the coexistence and interplay of these twofold interaction modes. Here, we introduce DMG-PPI, a dual-channel graph neural network framework that jointly models similarity and complementarity in PPI networks, extending prior heterophilous GNN concepts to explicitly disentangle these dual interaction modes. The core of the model consists of two parallel processing pathways: Alignment Message Passing (AMP), which aggregates information from proteins with similar features to capture interactions driven by shared structural patterns, and Divergence Message Passing (DMP), which emphasizes differences between proteins and identifies complementary features that may indicate physicochemical compatibility. The signals captured by AMP and DMP are integrated via an adaptive fusion strategy within each block, and the outputs of blocks are aggregated using the MixHop framework to encode higher-order interaction patterns. DMG-PPI substantially outperforms state-of-the-art methods on classical benchmark datasets, achieving a 7.19% improvement in Micro-F1 over the second-best method. Additionally, the dual-channel framework provides interpretable insights into key binding residues by identifying interfacial mechanisms. Overall, DMG-PPI serves as a powerful tool that reveals the mechanisms behind accurate PPI predictions and facilitates downstream biological analysis.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Substanzen
Autor:innen und Institutionen
- School of Modern Posts, Nanjing University of Posts and Telecommunications, Jiangsu, China.
- Department of Computer Science, University of Tsukuba, Ibaraki, Japan.
- School of Chinese Medicine, Hong Kong Baptist University, Kowloon, Hong Kong SAR, China.
- College of Computer Science and Electronic Engineering, Hunan University, Hunan, China.
- Yuelushan Laboratory, Changsha, China.
- College of Biology, Hunan University, Hunan, China.
ORCID-Profile: Tao Tang, Yuansheng Liu
Förderung
- National Nature Science Foundation of China
Zitieren
APA
Tang, T., Shen, T., Li, W., Chen, Y., Yuan, S., Liu, Y., Yang, X., & Luo, X. (2026). Dual-channel graph learning reveals similarity and complementarity in protein-protein interaction networks. PLOS Computational Biology, 22(5), e1013941. https://doi.org/10.1371/journal.pcbi.1013941
Vancouver
Tang T, Shen T, Li W, Chen Y, Yuan S, Liu Y, et al. Dual-channel graph learning reveals similarity and complementarity in protein-protein interaction networks. PLOS Computational Biology. 2026;22(5):e1013941. doi:10.1371/journal.pcbi.1013941
BibTeX
@article{Tang2026,
author = {Tang, Tao and Shen, Taiguang and Li, Weizhuo and Chen, Yangyang and Yuan, Sisi and Liu, Yuansheng and Yang, Xinyu and Luo, Xiao},
title = {Dual-channel graph learning reveals similarity and complementarity in protein-protein interaction networks},
journal = {PLOS Computational Biology},
year = {2026},
volume = {22},
number = {5},
pages = {e1013941},
doi = {10.1371/journal.pcbi.1013941}
}
EUVIMED