EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

BiGvCL: bipartite graph-based cross-domain contrastive learning model for the predicting drug-gene interactions

Shida He iD, Zixu Wang, Jing Li iD, Quan Zou iD, Feng Zhang

Briefings in Bioinformatics · 2026

DOI: 10.1093/bib/bbaf710 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Briefings in Bioinformatics, 2026. Verfasst von Shida He und 4 weiteren Autor:innen. Beteiligte Einrichtung: The Joint Innovation Center for Engineering in Medicine, Quzhou Affiliated Hospital of Wenzhou Medical University, Quzhou People's Hospital, No. 100,…. Hauptthemen laut MEDLINE: Computational Biology, Machine Learning. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 83 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, Zhejiang Provincial Natural Science Foundation of China, Municipal Government of Quzhou.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Drug-gene interactions (DGIs) influence the toxicity or ineffectiveness of the drug therapy and play an important role in elucidating drug mechanisms, predicting potential adverse effects, and facilitating precision medicine. Existing computational methods typically rely on chemical or genetic sequence features of drugs and genes, limiting their effectiveness for novel entities lacking explicit annotations. To address this, we propose BiGvCL, a framework that predicts DGIs exclusively based on network topology, requiring no explicit feature information for drugs or genes. BiGvCL introduces a lightweight graph attention mechanism (GATLite) to efficiently aggregate local neighborhood information. Additionally, we develop a gated graph convolutional network (GatedGCN) to explicitly learn high-order interactions between drugs and genes, further integrating contrastive learning to enhance the model’s generalizability. Comprehensive experiments on DrugBank and DGIdb datasets show that BiGvCL achieves competitive performance across all metrics compared with representative baselines. Cross-domain evaluations on OGB datasets further confirm its adaptability to heterogeneous biomedical networks. Ablation and hyperparameter analyses highlight the key contributions of contrastive and gated mechanisms, while case studies and molecular docking provide supporting evidence for the biological relevance of predictions. Collectively, while BiGvCL is constrained by its reliance on network topology and transductive learning paradigm, it demonstrates the potential of topology-based approaches for discovering novel drug-gene interactions, which may inform drug repurposing and precision medicine efforts.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Computational BiologyMachine LearningGraph Neural NetworksHumansPharmaceutical PreparationsAlgorithms

Stichwörter der Autor:innen

contrastive learningcross-domain knowledge transferdrug-gene interactionsgraph neural networkstransductive learning

Substanzen

Pharmaceutical Preparations

Autor:innen und Institutionen

  • The Joint Innovation Center for Engineering in Medicine, Quzhou Affiliated Hospital of Wenzhou Medical University, Quzhou People's Hospital, No. 100, Minjiang Avenue, Kecheng District, Quzhou, Zhejiang, 324000, China.
  • Department of Respiratory and Critical Care, Quzhou Affiliated Hospital of Wenzhou Medical University, No. 100, Minjiang Avenue, Kecheng District, Quzhou, Zhejiang, 324000, China.
  • Yangtze Delta Region Institute (Quzhou), University of Electronic Science and Technology of China, No. 1 Chengdian Road, Kecheng District, Quzhou, Zhejiang, 324000, China.
  • College of Computer Science and Electronic Engineering, Hunan University, Lushan South Road, Yuelu District, Changsha, Hunan, 410082, China.
  • Department of Microbiology, University of Hong Kong, Block T, Queen Mary Hospital, Pok Fu Lam Road, Pok Fu Lam, Hong Kong, 000000, China.

ORCID-Profile: Shida He, Jing Li, Quan Zou

Förderung

  • National Natural Science Foundation of China
  • Zhejiang Provincial Natural Science Foundation of China
  • Municipal Government of Quzhou

Zitieren

APA

He, S., Wang, Z., Li, J., Zou, Q., & Zhang, F. (2026). BiGvCL: bipartite graph-based cross-domain contrastive learning model for the predicting drug-gene interactions. Briefings in Bioinformatics. https://doi.org/10.1093/bib/bbaf710

Vancouver

He S, Wang Z, Li J, Zou Q, Zhang F. BiGvCL: bipartite graph-based cross-domain contrastive learning model for the predicting drug-gene interactions. Briefings in Bioinformatics. 2026. doi:10.1093/bib/bbaf710

BibTeX
@article{He2026,
  author = {He, Shida and Wang, Zixu and Li, Jing and Zou, Quan and Zhang, Feng},
  title = {BiGvCL: bipartite graph-based cross-domain contrastive learning model for the predicting drug-gene interactions},
  journal = {Briefings in Bioinformatics},
  year = {2026},
  doi = {10.1093/bib/bbaf710}
}
Kostenlos starten →