Graph Regularized Probabilistic Matrix Factorization for Drug-Drug Interactions Prediction

TL;DR

This paper introduces a novel graph-regularized probabilistic matrix factorization method for predicting drug-drug interactions, leveraging expert knowledge to improve accuracy in identifying adverse reactions.

Contribution

The paper proposes a new GRPMF method that integrates graph-based regularization into matrix factorization for DDI prediction, with an efficient optimization algorithm.

Findings

GRPMF outperforms state-of-the-art methods on DrugBank dataset.

Incorporating expert knowledge improves prediction accuracy.

The method effectively handles the non-convex optimization problem.

Abstract

Co-administration of two or more drugs simultaneously can result in adverse drug reactions. Identifying drug-drug interactions (DDIs) is necessary, especially for drug development and for repurposing old drugs. DDI prediction can be viewed as a matrix completion task, for which matrix factorization (MF) appears as a suitable solution. This paper presents a novel Graph Regularized Probabilistic Matrix Factorization (GRPMF) method, which incorporates expert knowledge through a novel graph-based regularization strategy within an MF framework. An efficient and sounded optimization algorithm is proposed to solve the resulting non-convex problem in an alternating fashion. The performance of the proposed method is evaluated through the DrugBank dataset, and comparisons are provided against state-of-the-art techniques. The results demonstrate the superior performance of GRPMF when compared to its counterparts.