2017/06/21 by S. Chtita, M. A. Ajana
Computer Science · Chemistry · Biochemistry, Genetics and Molecular Biology · #Computational Drug Discovery Methods #Synthesis and biological activity #Metabolomics and Mass Spectrometry Studies
paper · doi:10.48317/imist.prsm/morjchem-v6i4.9333
A Quantitative Structure–Activity Relationship (QSAR) study was performed to predict an anticancer activity in tumor cells of thirty-six 5-N-substituted-2-(substituted benzenesulphonyl) glutamines compounds using the electronic and topologic descriptors computed respectively, with ACD/ChemSketch and Gaussian 03W programs. The structures of all 36 compounds were optimized using the hybrid Density Functional Theory (DFT) at the B3LYP/6-31G(d) level of theory. In both approaches, 30 compounds were assigned as the training set and the rest as the test set. These compounds were analyzed by the Principal Components Analysis (PCA) method, a descendant Multiple Linear Regression (MLR), Multiple Nonlinear Regression (MNLR) analyses and an Artificial Neural Network (ANN). The robustness of the obtained models was assessed by leave-many-out cross-validation, and external validation through a test set. This study shows that the ANN has served marginally better to predict antitumor activity when compared with the results given by predictions made with MLR and MNLR