An sp3 hybridized carbon atom was used as a probe atom to generate steric (Lennard–Jones potential) field energies and a charge of +1 to generate electrostatic (Coulombic potential) field energies. A distance dependent dielectric constant of 1.00 was used. The steric and electrostatic fields were truncated at +30.00 kcal mol−1. The similarity indices descriptors were calculated using the same lattice box employed for CoMFA calculations, using sp3 carbon as a probe atom with +1 charge, +1 hydrophobicity and +1 H-bond donor and +1 H-bond acceptor properties. A partial least squares regression was used
to generate a linear INK1197 order relationship that correlates changes in the computed fields with changes in the corresponding experimental values of biological activity (pIC50) for the data set of ligands. Biological activity values of ligands
were used as dependent variables in a PLS statistical analysis.17 The column filtering value(s) was set to 2.0 kcal mol−1 to improve the signal-to-noise ratio by omitting those lattice points whose energy variations were below this threshold. Cross-validations were performed by the leave-one-out (LOO) procedure to determine the optimum number of components Doxorubicin concentration (ONC) and the coefficient q 2. The optimum number of components obtained is then used to derive the final QSAR model using all of the training set compounds with non-cross validation and to obtain the conventional correlation coefficient (r 2). To validate the CoMFA and CoMSIA derived models, the predictive ability for the test set of compounds (expressed as rpred2) was determined by using the following equation: rpred2=(SD−PRESS)/SD SD is the sum of the squared deviations between the biological activities of the test set
molecules and the mean activity of the training set compounds. PRESS is the sum of the squared deviation between the observed and the predicted activities of the test 4-Aminobutyrate aminotransferase set compounds. Since the statistical parameters were found to be the best for the model from the LOO method, it was employed for further predictions of the designed molecules. The 3D QSAR – CoMFA and CoMSIA analysis were carried out using small molecules like bezoxazol-5-yl acetic acid derivatives and 1,3-bis[4-(1H-bezimidazol-2-yl)-phenyl urea reported as potent inhibitors of heparanase9, 10 and 11 having precise IC50 value. A total of 43 molecules were used for derivation of model, these were divided into a training set of 33 molecules and test set of ten. The CoMFA and CoMSIA statistical analysis is summarized in Table 2. Statistical data shows qloo2 0.505 for CoMFA 0.540 for CoMSIA models, rncv2 of 0.972 and 0.988 for CoMFA and CoMSIA, respectively, which indicates a good internal predictive ability of the models. To test the predictive ability of the models, a test set of ten molecules excluded from the model derivation was used. The predictive correlation coefficient rpred2 of 0.556 for CoMFA and 0.