Quantitative Structure-Electrochemistry Relationship Study for Prediction of Half-Wave Reduction Potentials of Some Chlorinated Organic Compounds by Genetic Algorithm-Multiple Linear Regression

Abstract

Quantitative structure-electrochemistry relationship model has been used to predict and explain half-wave reduction potentials (E_1/2). This method allows for the prediction of E_1/2s in a variety of organic compounds based on their structures alone. Genetic algorithm-multiple linear regression (GA-MLR) was performed to build the model. The proposed methodology was validated using leave-one-out and leave-group-out cross validation using division of the available data set into training and test sets. The results illustrated that the linear techniques such as multiple linear regression combined with a successful variable selection procedure like GA are capable to generate an efficient quantitative structure-electrochemistry relationship model for predicting the E_1/2s of different compounds. A model with low prediction error and good correlation coefficient was obtained (R²_calibration = 0.923, R²_prediction = 0.940, Q²_LOO = 0.810, Q²_LGO = 0.803, R²_adj = 0.889, RMSEP = 0.203). This model was used for the prediction of the E_1/2 values of some organic compounds, which were not used in the modeling procedure.