Quantitative structure activity relationship of polyphenol derivatives as P-glycoprotein inhibitor

Abstract

Introduction and Background: P-glycoprotein (P-gp) is one of the membrane transporter proteins which can efflux drugs to the out of the cell and cause drug resistance. Therefore, designing of new compounds with P-gp inhibitory activity can reduce drug resistance. Aims: Our aim is to introduce quantitative structure activity relationship (QSAR) by Dragon descriptors and image-based QSAR models for predicting the p-glycoprotein inhibitory activity of polyphenol derivatives. Methods: The 2D-chemical structures and their P-gp inhibitory activity were taken from literature. Molecular descriptors were calculated using Dragon software. The pixels of images and their principal components (PCs) were calculated using MATLAB software. The most suitable descriptors and PCs were selected using stepwise regression. Principle component regression (PCR), multiple linear regression (MLR), artificial neural network (ANN) and support vector machine (SVM) approaches were used to develop QSAR models. Internal and external validations were applied to validate the QSAR models. Results: Five structural descriptors from Dragon software and six PCs from image analysis method were selected by stepwise regression for developing linear and non-linear models. Non-linear models i.e. ANN (with the correlation coefficient (R2) of 0.80 for test set) and SVM (with the R2 of 0.94 for test set) were chosen as the best for the established QSAR models by Dragon descriptors and image-based QSAR models, respectively. Discussion and Conclusion: The developed QSAR models were able to predict the P-gp inhibitory activity of the studied compounds with good accuracy. Non-linear models and rational training and test set selection methods can introduce better results for predicting the activity. The results of this study suggest that image based analysis can be economic and fast method for prediction of drug-like molecules without needing to commercial software.