Asymmetric Bagging and Feature Selection for ActivitiesPrediction of Drug Molecules

Activities of drug molecules can be predicted by QSAR (quantitative structure activity relationship) models, which overcome the disadvantage of high cost and long cycle by employing the traditional experimental method. With the fact that the number of drug molecules with positive activity is rather fewer than that with negatives, it is important to predict molecular activities considering such an unbalanced situation. Here, asymmetric bagging and feature selection are introduced into the problem and asymmetric bagging of support vector machines (AB-SVM) is proposed on predicting drug activities to treat the unbalanced problem. At the same time, the features extracted from the structures of drug molecules aspects prediction accuracy of QSAR models. Therefore, a novel algorithm named PRIFEAB is proposed, which applies an embedded feature selection method to remove redundant and irrelevant features for AB-SVM. Numerical experimental results on a data set of molecular activities show that AB-SVM improves the A UC values of molecular activities, and PRIFEAB with feature selection further helps to improve the prediction ability.