Machine Learning and Experimental Verification Identify Anti-Influenza Natural Products
International Journal of Molecular Sciences · 15.06.2026 · 27 (12) · 5399
DOI: 10.3390/ijms27125399 ↗Kurz gefasst
Fachartikel in der Zeitschrift International Journal of Molecular Sciences, online erschienen am 15.06.2026. Verfasst von Feifan Qiu und 9 weiteren Autor:innen. Beteiligte Einrichtung: College of Integrated Chinese and Western Medicine, Changchun University of Chinese Medicine, Changchun 130117, China. Hauptthemen laut MEDLINE: Antiviral Agents, Influenza A Virus, H1N1 Subtype, Machine Learning, Biological Products. Die Arbeit nennt 58 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, China National Natural Science Foundation.
Abstract (Originalsprache)
The influenza A virus (IAV) has been responsible for multiple seasonal epidemics and poses a pandemic threat, and the growing number of variant strains constitutes a persistent threat to humanity. This study aimed to identify anti-influenza compounds from a traditional Chinese medicine (TCM) monomer library using a machine learning approach, with calmodulin as a hypothesis-driven target. The antiviral efficacy of the compounds with the highest predicted binding scores from virtual screening was evaluated using integrated computational and experimental approaches. A pre-trained protein language model (ConPLex) was employed for virtual screening. Molecular docking was used to predict binding characteristics, and network pharmacology was applied to generate hypotheses on multi-target mechanisms. The cytotoxicity and anti-H1N1 activity of the selected compounds were assessed in vitro, followed by in vivo evaluation of survival, lung pathology, viral load, and inflammatory mediators in a lethal mouse infection model. Sodium deoxycholate (NaDC) and deoxycholic acid (DCA) were identified as promising lead compounds. Both exhibited dose-dependent inhibition of viral replication in vitro with low cytotoxicity. Treatment with NaDC and DCA significantly improved survival rates and reduced lung pathology in H1N1-infected mice. Treatment was associated with suppression of nuclear factor kappa-B (NF-κB) activation, reduced pro-inflammatory cytokines, and elevated interleukin-10 (IL-10) levels. Molecular docking predictions indicated that NaDC and DCA exhibit moderate binding affinity for calmodulin, with binding energies of −8.38 kcal/mol and −7.61 kcal/mol, respectively. Furthermore, network pharmacology analysis suggested that these compounds may modulate pathways related to viral infection, inflammation, and immune regulation. NaDC and DCA demonstrate anti-influenza activity both in vitro and in vivo, reducing viral replication and alleviating inflammatory lung injury. These findings position NaDC and DCA as promising lead compounds for anti-influenza drug development and provide a foundation for further mechanistic validation.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Substanzen
Autor:innen und Institutionen
- College of Integrated Chinese and Western Medicine, Changchun University of Chinese Medicine, Changchun 130117, China.
- State Key Laboratory of Pathogen and Biosecurity, Key Laboratory of Jilin Province for Zoonosis Prevention and Control, Changchun Veterinary Research Institute, Chinese Academy of Agricultural Sciences, Changchun 130122…
ORCID-Profile: Yueqi Wang
Förderung
- National Natural Science Foundation of China
- China National Natural Science Foundation
Zitieren
APA
Qiu, F., Wu, J., Cao, Y., Li, X., Wang, S., Xue, K., Wang, Y., Bu, Y., Shen, B., & Gao, Y. (2026). Machine Learning and Experimental Verification Identify Anti-Influenza Natural Products. International Journal of Molecular Sciences, 27(12), 5399. https://doi.org/10.3390/ijms27125399
Vancouver
Qiu F, Wu J, Cao Y, Li X, Wang S, Xue K, et al. Machine Learning and Experimental Verification Identify Anti-Influenza Natural Products. International Journal of Molecular Sciences. 2026;27(12):5399. doi:10.3390/ijms27125399
BibTeX
@article{Qiu2026,
author = {Qiu, Feifan and Wu, Jiajing and Cao, Yan and Li, Xuena and Wang, Shuo and Xue, Kun and Wang, Yueqi and Bu, Yizhou and Shen, Beilei and Gao, Yuwei},
title = {Machine Learning and Experimental Verification Identify Anti-Influenza Natural Products},
journal = {International Journal of Molecular Sciences},
year = {2026},
volume = {27},
number = {12},
pages = {5399},
doi = {10.3390/ijms27125399}
}
EUVIMED