QSRR Study of the Polychlorinated Biphenyl Congeners gainst the Retention Time Using the Multiple Linear Regression-Genetic Algorithms

摘要

The quantitative structure-retention relationship (QSRR) was applied to modeling and predicting the retention time of the polychlorinated biphenyl congeners. The suitable set of molecular descriptors was calculated and the important descriptors were selected with the aid of the genetic algorithm and stepwise multiple regression methods. A comparison between the attained results indicated the superiority of the genetic algorithm over the stepwise multiple regression method in the feature-selection. The simple GAMLR model with two selected descriptors was obtained. The models were validated using Leave-One-Out (LOO) and Leave-Group-Out (LGO) cross validation, and Y- randomization test. The statistical measures such as R2, Q2, and F value of this model were 0.986, 0.986 and 5881.823 respectively, revealing the credibility of the model.