KG-CNNDTI: a knowledge graph-enhanced prediction model for drug-target interactions and application in virtual screening of natural products against Alzheimer's disease.
10.1016/S1875-5364(25)60980-0
- Author:
Chengyuan YUE
1
;
Baiyu CHEN
1
;
Long CHEN
1
;
Le XIONG
1
;
Changda GONG
1
;
Ze WANG
1
;
Guixia LIU
1
;
Weihua LI
1
;
Rui WANG
2
;
Yun TANG
3
Author Information
1. Shanghai Frontiers Science Center of Optogenetic Techniques for Cell Metabolism, Shanghai Key Laboratory of New Drug Design, School of Pharmacy, East China University of Science and Technology, Shanghai 200237, China.
2. Shanghai Frontiers Science Center of Optogenetic Techniques for Cell Metabolism, Shanghai Key Laboratory of New Drug Design, School of Pharmacy, East China University of Science and Technology, Shanghai 200237, China. Electronic address: ruiwang@ecust.edu.cn.
3. Shanghai Frontiers Science Center of Optogenetic Techniques for Cell Metabolism, Shanghai Key Laboratory of New Drug Design, School of Pharmacy, East China University of Science and Technology, Shanghai 200237, China. Electronic address: ytang234@ecust.edu.cn.
- Publication Type:Journal Article
- Keywords:
Alzheimer’s disease;
Drug screening;
Drug-target interactions prediction;
Knowledge graph;
Natural products
- MeSH:
Alzheimer Disease/drug therapy*;
Biological Products/therapeutic use*;
Humans;
Neural Networks, Computer;
Machine Learning;
Drug Discovery/methods*;
Algorithms;
Drug Evaluation, Preclinical/methods*
- From:
Chinese Journal of Natural Medicines (English Ed.)
2025;23(11):1283-1292
- CountryChina
- Language:English
-
Abstract:
Accurate prediction of drug-target interactions (DTIs) plays a pivotal role in drug discovery, facilitating optimization of lead compounds, drug repurposing and elucidation of drug side effects. However, traditional DTI prediction methods are often limited by incomplete biological data and insufficient representation of protein features. In this study, we proposed KG-CNNDTI, a novel knowledge graph-enhanced framework for DTI prediction, which integrates heterogeneous biological information to improve model generalizability and predictive performance. The proposed model utilized protein embeddings derived from a biomedical knowledge graph via the Node2Vec algorithm, which were further enriched with contextualized sequence representations obtained from ProteinBERT. For compound representation, multiple molecular fingerprint schemes alongside the Uni-Mol pre-trained model were evaluated. The fused representations served as inputs to both classical machine learning models and a convolutional neural network-based predictor. Experimental evaluations across benchmark datasets demonstrated that KG-CNNDTI achieved superior performance compared to state-of-the-art methods, particularly in terms of Precision, Recall, F1-Score and area under the precision-recall curve (AUPR). Ablation analysis highlighted the substantial contribution of knowledge graph-derived features. Moreover, KG-CNNDTI was employed for virtual screening of natural products against Alzheimer's disease, resulting in 40 candidate compounds. 5 were supported by literature evidence, among which 3 were further validated in vitro assays.
