Partition Coefficient Prediction of a Large Set of Various Drugs and Poisons by a Genetic Algorithm and Artificial Neural Network

Quantum chemical calculations to reveal the relationship between the chemical structure and the fluorescence characteristics of phenylquinolinylethynes and phenylisoquinolinylethynes derivatives, and to predict their relative fluorescence intensity

In this paper the relationship between the chemical structure and fluorescence characteristics of 30 phenylquinolinylethyne (PhQE), and phenylisoquinolinylethyne (PhIE) derivatives compounds employing ab initio calculations have been elucidated. Quantum chemical calculations (6-31G) were carried out to obtain: the optimized geometry, energy levels, charges and dipole moments of these compounds, in the singlet (steady and excited states) and triplet states. The relationship between quantum chemical descriptors, and wavelength of maximum excitation and emission indicated that these two parameters have the most correlation with quantum chemical hardness (eta). Also, stokes shift has the most correlation with the square of difference between the maximum of positive charges in the singlet steady and singlet excited states. The quantitative structure-property relationship (QSPR) of PhQE and PhIE was studied for relative fluorescence intensity (RFI). The genetic algorithm (GA) was applied to select the variables that resulted in the best-fit models. After the variable selection, multiple linear regression (MLR) and support vector machine (SVM) were both utilized to construct linear and non-linear QSPR models, respectively. The SVM model demonstrated a better performance than that of the MLR model. The route mean square error (RMSE) in the training and the test sets for the SVM model was 0.195 and 0.324, and the correlation coefficients were 0.965 and 0.960, respectively, thus revealing the reliability of this model. The resulting data indicated that SVM could be used as a powerful modeling tool for QSPR studies. According to the best of our knowledge, this is the first research on QSPR studies to predict RFI for a series of PhQE and PhIE derivative compounds using SVM.

Exploring QSARs for antiviral activity of 4-alkylamino-6-(2-hydroxyethyl)-2-methylthiopyrimidines by support vector machine

The support vector machine, which is a novel algorithm from the machine learning community, was used to develop quantitative structure activity relationship models to predict the antiviral activity of 4-alkylamino-6-(2-hydroxyethyl)-2-methylthiopyrimidines. The genetic algorithm was employed to select the variables that resulted in the best-fitted models. A comparison between the obtained results using support vector machine with those of multiple linear regression revealed that support vector machine model was much better than multiple linear regression. The root mean square errors of the training set and the test set for support vector machine model were calculated to be 0.102 and 0.205, and the correlation coefficients (r2) were 0.956 and 0.852, respectively. Furthermore, the obtained statistical parameter of leave-one-out (LOO) and leave-group-out (LGO) cross-validation test on support vector machine model were 0.893 and 0.881, respectively, which prove the reliability of this model. The results suggest that branching, volume and lipophilicity are the main independent factors contributing to the antiviral activities of the studied compounds.