Computational screening and QSAR study on a series of theophylline derivatives as aldh1a1 inhibitors

Abstract

In the present study, we explored a series of molecules with anticancer activity, so that qualitative and quantitative studies of the structure-activity relationship (SAR/QSAR) were performed on seventeen theophylline derivatives. These are inhibitors of ALDH1A1. The present study shows the importance of quantum chemical descriptors, constitutional descriptors and hydrophobicity to develop a better QSAR model, whose studied descriptors are LogP, MW, Pol, MR, S, V, HE, DM, EHOMO and ELUMO.

A multiple linear regression (MLR) and artificial neural networks (ANN) procedure was used to design the relationships between molecular descriptors and the inhibition of ALDH1A1 by theophylline derivatives. The validation and good quality of the QSAR model are confirmed by a strong correlation between experimental and predicted activity.