DFT-based QSAR study of N-hydroxyfurylacrylamide and its derivatives,principal inhibitors of histone deacetylase

A data set of 38 n-hydroxyfurylacrylamide derivatives,principal inhibitors of histone deacetylase (hdac) is used for quantitative structure-activity relationship (qsar) study. the physicochemical properties of the tested compounds have been described by a total of six descriptors comprising four quantum chemical descriptors and two molecular descriptors. molecules geometries optimization is performed using firstly dft method with b3lyp/6-31g (d) level,to generate descriptors based on electronic properties,and secondly molecular mechanics method with mm+ force field to generate molecular descriptors. qsar models were constructed using multiple linear regression method (mlr) and artificial neural network (ann) techniques. descriptors based on structural and electronic properties were shown to be important in classifying the compounds. good correlations are shown both for mlr and ann towards hdac inhibitory activities (0.80 and 0.98,respectively). so,ann model established with ann techniques considering the relevant descriptors selected by mlr,is statistically significant and show very good stability towards data variation with leave-one-out (loo) test,the most convenient procedure of cross validation method (rloo =0.77).