Quantitative structure activity relationship indanylacetic acid derivatives models based on a novel machine learning method

Peijian Zhang, Hongzong Si, Yu

Abstract

Quantitative structure activity relationship (QSAR) models for prediction of the EC50 of the peroxisome proliferator-activated receptors (PPARs) have been developed basis on the linear heuristic method(HM). Molecular descriptors were used to represent the characteristics of compounds. HM was used to pre-select the whole descriptor sets and to build the linear model. The new compounds were designed according to the QSAR models. The same descriptors were applied in the model and the satisfied EC50 values were obtained for the designed compounds. The selected descriptors will help for new drug design and the models are available for predicting the EC50 of the new drugs.