In silico studies of novel scaffold of thiazolidin-4-one derivatives as anti-Toxoplasma gondii agents by 2D/3D-QSAR, molecular docking, and molecular dynamics simulations

Abstract

Toxoplasma gondii is an obligate intracellular protozoa that can infect a wide variety of warm-blooded animals and humans. It was claimed that novel anti-Toxoplasma gondii agents were optimized as potential drug candidates, designed and created as significant agents. In this work, molecular modeling studies, including CoMFA, CoMFA-RF, CoMSIA, and HQSAR were performed on a set of 59 thiazolidin-4-one derivatives as anti-T. gondii agents. The statistical qualities of generating models were justified by internal and external validation, i.e., cross-validated correlation coefficient (q2), non-cross-validated correlation coefficient ($$ {r}_{ncv}^2 $$) and predicted correlation coefficient ($$ {r}_{pred}^2 $$), respectively. The CoMFA (q2, 0.897;$$ \kern0.5em {r}_{ncv}^2 $$, 0.933; $$ {r}_{pred}^2 $$, 0.938), CoMFA-RF (q2, 0.900;$$ \kern0.5em {r}_{ncv}^2 $$, 0.935; $$ {r}_{pred}^2 $$, 0.998), CoMSIA (q2, 0.910;$$ {r}_{ncv}^2 $$, 0.950; $$ {r}_{pred}^2 $$, 0.998), and HQSAR models (q2, 0.924;$$ {r}_{ncv}^2 $$, 0.953; $$ {r}_{pred}^2 $$, 0.995) for training and test set yielded significant statistical results. Therefore, these QSAR models were excellent, robust, and had better predictive capability. Contour maps of the QSAR models were generated and validated by molecular dynamics simulation-assisted molecular docking study. The final QSAR models could be useful for the design and development of novel potent anti-T. gondii agents.