A simple approach to the prediction of electrophoretic mobilities of diverse organic acids

Based on the principle of separation by capillary electrophoresis, three simple molecular structural descriptors, the number of acid groups (NG), number of atoms (NA) and molecular weight (MW) were employed to model the molecular structure-electrophoretic mobility relationship with the general form of mu(o) = aNG/(bNA(k) + cMW(k)), where the value of k ranges from 1/3 to 2/3 depending on the magnitude of molecular weight. The absolute electrophoretic mobilities (mu(o)) of three diverse data sets including carboxylic acids, sulfonic acids and peptides were investigated. Each data set was randomly divided into a training set for model generation and a test set for prediction test. The absolute average relative deviations (AARD and AARD(p)) of the training set and the test set are 7.21% and 7.53% for data set 1, 1.99% and 1.96% for data set 2, 3.25% and 3.95% for data set 3, respectively. Compared to the quantitative structure-mobility relationship (QSMR) models reported in the literature, the approach proposed in this study is much simpler in terms of the calculation of structural descriptors and the interpretation of the physicochemical meaning of the models. (c) 2006 Elsevier B.V. All rights reserved.