Identification of SVM-based classification model, synthesis and evaluation of prenylated flavonoids as vasorelaxant agents.

Support vector machine (SVM) was applied to predict vasorelaxation effect of different structural molecules. A good classification model had been established, and the accuracy in prediction for the training, test, and overall datasets was 93.0%, 82.6%, and 89.5%, respectively. Furthermore, the model was used to predict the activity of a series of prenylated flavonoids. According to the estimated result, eleven molecules 1-11 were selected and synthesized. Their vasodilatory activities were determined experimentally in rat aorta rings that were pretreated with phenylephrine (PE). Structure-activity relationship (SAR) analysis revealed that flavanone derivatives showed the most potent activities, while flavone and chalcone derivatives exhibited medium activities.