Multi-view Graph Contrastive Representation Learning for Drug-Drug Interaction Prediction

TL;DR

This paper introduces MIRACLE, a novel multi-view graph contrastive learning method that effectively captures complex drug interactions by integrating molecular structures and interaction networks, leading to improved DDI prediction accuracy.

Contribution

MIRACLE is the first to combine multi-view graph contrastive learning with drug molecular and interaction data for DDI prediction.

Findings

MIRACLE outperforms existing models on multiple datasets.

The method effectively captures inter-view and intra-view drug information.

Unsupervised contrastive learning enhances model performance.

Abstract

Drug-drug interaction(DDI) prediction is an important task in the medical health machine learning community. This study presents a new method, multi-view graph contrastive representation learning for drug-drug interaction prediction, MIRACLE for brevity, to capture inter-view molecule structure and intra-view interactions between molecules simultaneously. MIRACLE treats a DDI network as a multi-view graph where each node in the interaction graph itself is a drug molecular graph instance. We use GCNs and bond-aware attentive message passing networks to encode DDI relationships and drug molecular graphs in the MIRACLE learning stage, respectively. Also, we propose a novel unsupervised contrastive learning component to balance and integrate the multi-view information. Comprehensive experiments on multiple real datasets show that MIRACLE outperforms the state-of-the-art DDI prediction models consistently.