Machine learning-guided discovery and preliminary biological evaluation of novel GSK-3β inhibitory scaffolds with anti-inflammatory potential
Journal of Computer-Aided Molecular Design · 13.11.2026 · 40 (1)
DOI: 10.1007/s10822-026-00974-4 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Computer-Aided Molecular Design, online erschienen am 13.11.2026. Verfasst von Shiqi Cheng und 9 weiteren Autor:innen. Beteiligte Einrichtung: Department of Pharmacy, Daping Hospital, Army Medical University, Chongqing, 400042, China. Hauptthemen laut MEDLINE: Glycogen Synthase Kinase 3 beta, Machine Learning, Anti-Inflammatory Agents, Protein Kinase Inhibitors. Die Arbeit nennt 54 Literaturangaben. Gefördert von: the Natural Science Foundation of Chongqing, China, Chongqing Natural Science Foundation.
Abstract (Originalsprache)
Glycogen synthase kinase-3β (GSK-3β) is a multifunctional serine/threonine kinase implicated in neurodegeneration, cancer, metabolic disorders, and inflammatory diseases. The discovery of structurally novel GSK-3β inhibitors remains important for developing new chemical probes and therapeutic starting points. In this study, we established a machine learning-assisted computational screening workflow for the identification of potential GSK-3β inhibitors. A Random Forest (RF) classification model was trained using a curated dataset of reported GSK-3β-related compounds collected from 2016 to 2024. The optimized RF model showed good predictive performance, with an accuracy of 0.89 and an AUC-ROC of 0.91 ± 0.02, supporting its use as an initial prioritization tool for virtual screening. The trained classifier was applied to the Asinex compound library, and prioritized compounds were further evaluated through physicochemical and ADMET filtering, molecular docking, and molecular dynamics simulations. This integrated workflow led to the selection of two representative candidates, RF03 and RF05, for experimental validation. In vitro enzymatic assays confirmed moderate GSK-3β inhibitory activity, with IC₅₀ values of 85.04 µM for RF03 and 58.73 µM for RF05. Among them, RF05, which contains a non-classical and sterically hindered aromatic ester scaffold, also reduced the production of pro-inflammatory cytokines, including TNF-α and IL-6, while maintaining low cytotoxicity under the tested conditions. Overall, this study demonstrates the feasibility of combining machine learning-based prioritization with structure-based modeling and biological evaluation to identify previously underexplored GSK-3β inhibitory scaffolds. Although RF05 requires further optimization to improve potency, selectivity, and pharmacological properties, it provides a useful starting point for the development of next-generation GSK-3β-targeted anti-inflammatory agents.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Substanzen
Autor:innen und Institutionen
- Department of Pharmacy, Daping Hospital, Army Medical University, Chongqing, 400042, China.
- Medical Research Institute, Southwest University, Chongqing, 400715, China.
- College of Chemistry and Chemical Engineering, Chongqing University of Science and Technology, Chongqing, 401331, China.
- Medical Research Institute, Southwest University, Chongqing, 400715, China. cysylq@swu.edu.cn.
- College of Chemistry and Chemical Engineering, Chongqing University of Science and Technology, Chongqing, 401331, China. peishuchen928@163.com.
- Department of Pharmacy, Daping Hospital, Army Medical University, Chongqing, 400042, China. liuyao@tmmu.edu.cn.
Förderung
- the Natural Science Foundation of Chongqing, China
- Chongqing Natural Science Foundation
Zitieren
APA
Cheng, S., Zhang, W., Zhou, J., Su, Y., Li, J., Peng, D., Liu, F., Yang, L., Pei, S., & Liu, Y. (2026). Machine learning-guided discovery and preliminary biological evaluation of novel GSK-3β inhibitory scaffolds with anti-inflammatory potential. Journal of Computer-Aided Molecular Design, 40(1). https://doi.org/10.1007/s10822-026-00974-4
Vancouver
Cheng S, Zhang W, Zhou J, Su Y, Li J, Peng D, et al. Machine learning-guided discovery and preliminary biological evaluation of novel GSK-3β inhibitory scaffolds with anti-inflammatory potential. Journal of Computer-Aided Molecular Design. 2026;40(1). doi:10.1007/s10822-026-00974-4
BibTeX
@article{Cheng2026,
author = {Cheng, Shiqi and Zhang, Wenyuan and Zhou, Jinjing and Su, Yang and Li, Jiaming and Peng, Dan and Liu, Fang and Yang, Liqun and Pei, Shuchen and Liu, Yao},
title = {Machine learning-guided discovery and preliminary biological evaluation of novel GSK-3β inhibitory scaffolds with anti-inflammatory potential},
journal = {Journal of Computer-Aided Molecular Design},
year = {2026},
volume = {40},
number = {1},
doi = {10.1007/s10822-026-00974-4}
}
EUVIMED