Link Mining for Kernel-based Compound-Protein Interaction Predictions Using a Chemogenomics Approach

TL;DR

This paper enhances kernel-based chemogenomics virtual screening for compound-protein interactions by combining link mining techniques, resulting in improved prediction accuracy with minimal additional computational cost.

Contribution

The study introduces a link mining-based enhancement to the pairwise kernel method, boosting prediction accuracy in chemogenomics virtual screening.

Findings

Improved AUPR from 0.425 to 0.562 using the proposed method.

Calculation time increased only slightly compared to conventional methods.

Enhanced prediction of new compound-protein interactions.

Abstract

Virtual screening (VS) is widely used during computational drug discovery to reduce costs. Chemogenomics-based virtual screening (CGBVS) can be used to predict new compound-protein interactions (CPIs) from known CPI network data using several methods, including machine learning and data mining. Although CGBVS facilitates highly efficient and accurate CPI prediction, it has poor performance for prediction of new compounds for which CPIs are unknown. The pairwise kernel method (PKM) is a state-of-the-art CGBVS method and shows high accuracy for prediction of new compounds. In this study, on the basis of link mining, we improved the PKM by combining link indicator kernel (LIK) and chemical similarity and evaluated the accuracy of these methods. The proposed method obtained an average area under the precision-recall curve (AUPR) value of 0.562, which was higher than that achieved by the conventional Gaussian interaction profile (GIP) method (0.425), and the calculation time was only increased by a few percent.