In silico study on anticonvulsant activity of isoxazole and thiazole derivatives active in subcutaneous pentylenetetrazole animal model

• Main Authors: Adedirin Oluwaseye, Adamu Uzairu, Gideon A. Shallangwa, Stephen E. Abechi
• Format: Article
• Language: English
• Published: Elsevier 2020-01-01
• Series: Journal of King Saud University: Science
• Online Access: http://www.sciencedirect.com/science/article/pii/S1018364718302775

Quantitative structure-activity relationship (QSAR) and molecular docking studies have been done on 28 isoxazole and thiazole derivatives with anticonvulsant activity in subcutaneous pentylenetetrazole animal model. Physicochemical parameters obtained from PaDEL-Descriptors were utilized in the study. Parametric semi-empirical quantum technique PM3 available in Spartan 14 program was used to optimize the molecular structure of the dataset compounds. Genetic function algorithm, Modified-K-Medoid clustering, correlation analysis and multiple linear regressions were used to search for the best QSAR model. The model obtained had good statistical parameters (LOF = 0.056, R2 = 0.975, Q2 = 0.959, F3,15 = 198.058, R2(Pred.) = 0.761, PRESS = 0.058, SEE = 0.062 and cR2P = 0.887) and can be utilized to predict the anticonvulsant activity of compounds that are within its chemical space. Molecular docking analysis showed that the studied compounds had a better binding affinity for γ-aminobutyrate aminotransferase than vigabatrin which is a known inhibitor of γ-aminobutyrate aminotransferase. Keywords: QSAR, Molecular docking, Subcutaneous pentylenetetrazole, Genetic function algorithm