In this study, Hybrid Docking and quantitative structure-activity relationship (QSAR) methods were used to investigate the binding of 72 inhibitors to Glycogen synthase kinase-3 (GSK-3). Inhibitors of GSK-3 are increasingly being used as an essential component in the treatment of type 2 diabetes. The inhibitors were first examined for their ability to bind to the X-ray structure of the biological target GSK-3 using docking studies. Key interactions for ligand binding into the receptor active site were identified, which shared common features with those found for other known inhibitors, thus strengthening the results of this study. Modeling was then performed using multiple linear regression (MLR) and support vector machine (SVM). The robustness and the predictive performance of the developed models were tested using both the internal and external statistical validation (test set) of ten compounds, Kennard-Stones algorithm chosen out of 72 compounds. The SVM model with optimal parameters C of 30, γ of 0.55, and ε of 0.05 has the R2 (0.83, 0. 68) and RMS errors (0.083, 0.625) for the training and test sets, respectively, which are better than the MLR method (R2=0.72, 0.77 and RMS error=0.449, 0.----). The goal was to find novel chemical scaffolds as potential GSK-3 inhibitors by virtual screening using the model, which was based on 85% similarity with Benzamide, Pyrazolopyrimidine, and β-phenylalanine derivatives from the PubChem database. Furthermore, in silico-predicted ADME properties were investigated for the most promising molecules. The outcome of this investigation sheds light on the molecular characteristics of the binding of analogs to GSK-3 and identifies new possible inhibitors that have the potential to be developed into drugs. This significant contribution can aid in the design and optimization of therapeutic strategies against type 2 diabetes.
