Anti-SARS-CoV-2 drugs discovery by combining semantic information with knowledge graph structural information
10.7644/j.issn.1674-9960.2025.07.003
- VernacularTitle:融合语义信息与知识图谱结构信息的抗新型冠状病毒药物发现研究
- Author:
Jiaxin ZHOU
1
;
Yin ZHANG
Author Information
1. 军事科学院军事医学研究院,北京 100850
- Keywords:
knowledge graph completion;
knowledge graph embedding;
SARS-CoV-2;
drug discovery;
link prediction
- From:
Military Medical Sciences
2025;49(7):494-503
- CountryChina
- Language:Chinese
-
Abstract:
Objective To propose a knowledge graph embedding model that can help discover potential anti-SARS-CoV-2 drugs from approved drugs by combining semantic information with knowledge graph structural information.Methods Potential therapeutic drugs were predicted by using the head entity prediction task in knowledge graph completion.Results Six potential drugs were predicted,including naratriptan,sumatriptan,colchicine,doxorubicin,diphenhydramine and hydrocortisone.Conclusion The combination of semantic information and knowledge graphstructural information can enhance the representation capability of a knowledge graph embedding model,and provide a novel approach to research on anti-SARS-CoV-2 drug discovery.
