Measurement and ANN prediction of pH-dependent solubility of nitrogen-heterocyclic compounds

摘要: Based on the solubility of 25 nitrogen-heterocyclic compounds (NHCs) measured by saturation shake-flask method, artificial neural network (ANN) was employed to study quantitative relationship between structure and pH-dependent NHCs. With genetic algorithm-multivariate linear regression (GA-MLR) approach, five out 1497 molecular descriptors computed Dragon software were selected describe structures Using as well pH partial charge nitrogen atom NHCs (QN) inputs ANN, a structure–property (QSPR) model without using Henderson–Hasselbalch (HH) equation successfully developed predict aqueous in different water solutions. The prediction performed with an absolute average relative deviation (AARD) 5.9%, while HH approach gave AARD 36.9% for same It found that QN played very important role description and, QN, ANN became potential tool