Accurate quantitative structure–property relationship model of mobilities of peptides in capillary zone electrophoresis

Abstract

The aim of this work was to predict electrophoretic mobilities of peptides in capillary zone electrophoresis (CZE) using the linear heuristic method (HM) and a nonlinear radial basis function neural network (RBFNN). Two data sets, consisting of 125 peptides ranging in size between 2 and 14 amino acids and 58 peptides ranging in size between 2 and 39 amino acids, are researched to test applicability of the QSPR methods. In this study, the root mean squared (RMS) errors of the training set, the test set and the whole set of data set 1 are 1.3766, 1.5608 and 1.4157 and the correlation coefficients (R²) are 0.9740, 0.9671 and 0.9724 predicted by RBFNN, respectively. While the RMS errors of the training set, the test set and the whole set of data set 2 are 0.6279, 0.8145 and 0.6673 and the correlation coefficients (R²) are 0.9773, 0.9489 and 0.9732, respectively. So the Offord charge-over-mass term (Q/M^2/3) combined with descriptors calculated by CODESSA represents the structural features of the peptides appropriately. The electrophoretic mobilities of peptides can be accurately predicted by the linear and nonlinear model. Furthermore, the results of nonlinear model are closer to the experimental data than those of linear model.