EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

MGACL: Prediction Drug–Protein Interaction Based on Meta-Graph Association-Aware Contrastive Learning

Pinglu Zhang, Peng Lin iD, Dehai Li iD, Wanchun Wang, Xin Qi, Jing Li, Jianshe Xiong

Biomolecules · 2024

DOI: 10.3390/biom14101267 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Biomolecules, 2024. Verfasst von Pinglu Zhang und 6 weiteren Autor:innen. Beteiligte Einrichtung: Faculty of Information Science and Engineering, Ocean University of China, Qingdao 266003, China. Hauptthemen laut MEDLINE: Drug Discovery. Die Arbeit wurde bisher 7-mal zitiert (Crossref) und nennt 47 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Shandong Provincial Natural Science Foundation, Hainan Provincial Joint Project of Sanya Yazhou Bay Science and Technology City.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The identification of drug–target interaction (DTI) is crucial for drug discovery. However, how to reduce the graph neural network’s false positives due to its bias and negative transfer in the original bipartite graph remains to be clarified. Considering that the impact of heterogeneous auxiliary information on DTI varies depending on the drug and target, we established an adaptive enhanced personalized meta-knowledge transfer network named Meta Graph Association-Aware Contrastive Learning (MGACL), which can transfer personalized heterogeneous auxiliary information from different nodes and reduce data bias. Meanwhile, we propose a novel DTI association-aware contrastive learning strategy that aligns high-frequency drug representations with learned auxiliary graph representations to prevent negative transfer. Our study improves the DTI prediction performance by about 3%, evaluated by analyzing the area under the curve (AUC) and area under the precision–recall curve (AUPRC) compared with existing methods, which is more conducive to accurately identifying drug targets for the development of new drugs.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Drug DiscoveryNeural Networks, ComputerHumansMachine LearningProteinsAlgorithmsPharmaceutical PreparationsArea Under Curve

Stichwörter der Autor:innen

contrastive learningdrug–target interactiongraph neural networksheterogeneous graph representation

Substanzen

ProteinsPharmaceutical Preparations

Autor:innen und Institutionen

  • Faculty of Information Science and Engineering, Ocean University of China, Qingdao 266003, China.
  • Key Laboratory of Marine Drugs, Chinese Ministry of Education, School of Medicine and Pharmacy, Ocean University of China, Qingdao 266003, China.

ORCID-Profile: Peng Lin, Dehai Li

Förderung

  • Shandong Provincial Natural Science Foundation
  • Hainan Provincial Joint Project of Sanya Yazhou Bay Science and Technology City

Zitieren

APA

Zhang, P., Lin, P., Li, D., Wang, W., Qi, X., Li, J., & Xiong, J. (2024). MGACL: Prediction Drug–Protein Interaction Based on Meta-Graph Association-Aware Contrastive Learning. Biomolecules. https://doi.org/10.3390/biom14101267

Vancouver

Zhang P, Lin P, Li D, Wang W, Qi X, Li J, et al. MGACL: Prediction Drug–Protein Interaction Based on Meta-Graph Association-Aware Contrastive Learning. Biomolecules. 2024. doi:10.3390/biom14101267

BibTeX
@article{Zhang2024,
  author = {Zhang, Pinglu and Lin, Peng and Li, Dehai and Wang, Wanchun and Qi, Xin and Li, Jing and Xiong, Jianshe},
  title = {MGACL: Prediction Drug–Protein Interaction Based on Meta-Graph Association-Aware Contrastive Learning},
  journal = {Biomolecules},
  year = {2024},
  doi = {10.3390/biom14101267}
}
Kostenlos starten →