Prediction of photolysis half-lives of dihydroindolizines by genetic algorithm-multiple linear regression (GA-MLR)

Abstract

Quantitative structure–property relationship study was carried out for the prediction of photolysis half-lives of dihydroindolizines. Genetic algorithm variable selection was used in the multiple linear regression modeling. Four descriptors including topological (AAC), 2D-Autocorrelations (GATS7m), 3D-MoRSE (Mor19p), and GETAWAY (HATS3p) descriptors were selected by the algorithm, revealing that a hybrid of topologic, geometric, and electronic features of the molecule affects photolysis half-life of studied photochromic compounds. The results showed that the developed model provides statistically significant predictions of photolysis half-lives of dihydroindolizines. Copyright © 2016 John Wiley & Sons, Ltd.