QSAR Model of Hydroxy- or Methoxy-substituted Benzaldoximes and Benzaldehyde-o-alkyloximes as Tyrosinase Inhibitors

Quantitative structure-activity relationship (QSAR) study on the tyrosinase inhibition activities of hydroxy- or methoxy-substituted benzaldoximes and benzaldehyde-O-alkyloximes was performed using support vector machines (SVM) method. The predictive power of the models was verified with the leave one out cross validation test and independent test methods. The cross validation squared correlation coefficient value for optimal SVM model was 0.6880. Compared with stepwise multiple linear regression and back propagation artificial neural network models, the SVM model was the most powerful with a square of predictive correlation coefficient of 0.6117 for the test set.