Molecular dynamics-guided receptor-dependent 4D-QSAR studies of HDACs inhibitors

Histone deacetylases (HDACs) were highlighted as a novel category of anticancer targets. Several HDACs inhibitors were approved for therapeutic use in cancer treatment. Comparatively, receptor-dependent 4D-QSAR, LQTA-QSAR, is a new approach which generates conformational ensemble profiles of compounds by molecular dynamics simulations at binding site of enzyme. This work describes a receptor-dependent 4D-QSAR studies on hydroxamate-based HDACs inhibitors. The 4D-QSAR model was generated by multiple linear regression method of QSARINS. Leave-N-out cross-validation (LNO) and Y-randomization were performed to analysis of the independent test set and to verify the robustness of the model. Best 4D-QSAR model showed the following statistics: R-2 = 0.8117, Q(LOO)(2) = 0.6881, Q(LNO)(2) = 0.6830, R-Pred(2) = 0.884. The results may be used for further virtual screening and design for novel HDACs inhibitors. [GRAPHICS] .

Automated summary

A receptor-dependent 4D-QSAR model on hydroxamate-based HDACs inhibitors was studied, with robustness confirmed through cross-validation and randomization. The best model showed high statistical values, suitable for further drug screening and design purposes.