| Abstract
| - A method for predicting log P values for a diverse set of 1870 organic molecules has been developed basedon atom-type electrotopological-state (E-state) indices and neural network modeling. An extended set ofE-state indices, which included specific indices with a more detailed description of amino, carbonyl, andhydroxy groups, was used in the current study. For the training set of 1754 molecules the squared correlationcoefficient and root-mean-squared error were r2 = 0.90 and RMSLOO = 0.46, respectively. Structuralparameters which included molecular weight and 38 atom-type E-state indices were used as the inputs in39-5-1 artificial neural networks. The results from multilinear regression analysis were r2 = 0.87 and RMSLOO= 0.55, respectively. For a test set of 35 nucleosides, 12 nucleoside bases, 19 drug compounds, and 50general organic compounds (n = 116) not included in the training set, a predictive r2 = 0.94 and RMS =0.41 were calculated by artificial neural networks. The results for the same set by multilinear regressionwere r2 = 0.86 and RMS = 0.72. The improved prediction ability of artificial neural networks can beattributed to the nonlinear properties of this method that allowed the detection of high-order relationshipsbetween E-state indices and the n-octanol/water partition coefficient. The present approach was found to bean accurate and fast method that can be used for the reliable estimation of log P values for even the mostcomplex structures.
|
