High-dimensional QSAR classification model for anti-hepatitis C virus activity of thiourea derivatives based on the sparse logistic regression model with a bridge penalty

Abstract

This study addresses the problem of the high-dimensionality of quantitative structure-activity relationship (QSAR) classification modeling. A new selection of descriptors that truly affect biological activity and a QSAR classification model estimation method are proposed by combining the sparse logistic regression model with a bridge penalty for classifying the anti-hepatitis C virus activity of thiourea derivatives. Compared to other commonly used sparse methods, the proposed method shows superior results in terms of classification accuracy and model interpretation.