Use of Artificial Neural Network for a QSAR Study on Neurotrophic Activities of N-p-Tolyl/phenylsulfonyl L-Amino Acid Thiolester Derivatives
Jin Luo, Jiwei Hu, Liya Fu, Chun Liu, Xiaofei Jin · Procedia Engineering · 2011
The objective of the present work was to use artificial neural network to study the quantitative structure-activity relationship (QSAR) of the protective effects of N-p-tolyl/phenylsulfonyl L-amino acid thiolester derivatives on anoxic damage of rat pheochromocytoma (PC12) cells. Five molecular parameters of these target compounds, including heat of formation, total energy, dipole moment, the energy of the highest occupied molecular orbital and the energy of the lowest unoccupied molecular orbital, were calculated with the PM6 semi-empirical quantum mechanical method. A multilayer feed-forward (MLFF) network with back-propagation (BP) learning was employed in the present work with the molecular parameters as inputs and neurotrophic activities as outputs. Results showed that the neural network can provide a good prediction of neurotrophic activity and may be useful for predicting the bioactivity of new compounds of similar class.