Multiple linear regression and artificial neural network retention prediction models for ginsenosides on a polyamine-bonded stationary phase in hydrophilic interaction chromatography

The development of retention prediction models for the seven ginsenosides Rf, Rg1, Rd, Re, Rc, Rb2, and Rb1 on a polyamine-bonded stationary phase in hydrophilic interaction chromatography (HILIC) is presented. The models were derived using multiple linear regression (MLR) and artificial neural network (ANN) using the logarithm of the retention factor (log k) as the dependent variable for four temperature conditions (0, 10, 25, and 40 degrees C). Using stepwise MLR, the retention of the analytes in all the temperature conditions was satisfactorily described by a two-predictor model wherein the predictors were the percentage of ACN (%ACN) in the mobile phase and local dipole index (LDI) of the compounds. These predictors account for the contribution of the solute-related variable (LDI) and the influence of the mobile phase composition (%ACN) on the retention behavior of the ginsenosides. A comparison of the models derived from both MLR and ANN revealed that the trained ANNs showed better predictive abilities than the MLR models in all temperature conditions as demonstrated by their higher R(2) values for both training and test sets and lower average percentage deviation of the predicted log k from the observed log k of the test compounds. The ANN models also showed excellent performance when applied to the prediction of the seven ginsenosides in different sample matrices.