Quantitative Structure-Activity Relationship Modeling of the Antioxidant Activity of Some Plant Compounds on Graph of Molecular Descriptors

Abstract

Plants synthesize antioxidant compounds as a defense mechanism against reactive oxygen species. Recently, plant-derived antioxidantcompounds have attracted attention due to the increasing consumer awareness in the heath industry. However, traditional methods formeasuring the antioxidant activity of these compounds are time-consuming and costly. Therefore, our study constructed a quantitativestructure-activity relationship (QSAR) model that can predict antioxidant activity using graph convolutional networks (GCN) from plantstructural data. The accuracy (Acc) of the model reached 0.6 and the loss reached 0.03. Although with lower accuracy than previouslyreported QSAR models, our model showed the possibility of predicting DPPH antioxidant activity in a wide range of plant compounds(phenolics, polyphenols, vitamins, etc.) based on their graph structure.