Towards Interpretable Drug-Drug Interaction Prediction: A Graph-Based Approach with Molecular and Network-Level Explanations

2025-07-12

AI Summary

Existing DDI prediction methods suffer from two key limitations: (1) neglecting contextual dependencies between drug pairs, and (2) difficulty in jointly modeling molecular structures and biological networks to yield mechanistic interpretability. To address these, we propose a unified drug-pair modeling framework that treats each drug pair as a holistic entity. Our approach enables multi-scale representation learning via context-aware subgraph pooling and attention-guided influence pooling, and further enhances interpretability and representation diversity through mutual information minimization regularization. Technically, the framework integrates local subgraph extraction, hierarchical interaction graph construction, knowledge graph embedding, and graph neural network (GNN) techniques. Extensive experiments on multiple standard DDI benchmark datasets demonstrate significant improvements over state-of-the-art methods. Ablation studies and embedding visualizations validate the effectiveness of individual components and provide biologically meaningful insights into underlying DDI mechanisms.

Technology Category

• Knowledge Representation and Reasoning: Diagnosis and Abductive Reasoning
• Machine Learning: Graph-based Machine Learning
• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community

Application Category

• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Drug-drug interactions (DDIs) represent a critical challenge in pharmacology, often leading to adverse drug reactions with significant implications for patient safety and healthcare outcomes. While graph-based methods have achieved strong predictive performance, most approaches treat drug pairs independently, overlooking the complex, context-dependent interactions unique to drug pairs. Additionally, these models struggle to integrate biological interaction networks and molecular-level structures to provide meaningful mechanistic insights. In this study, we propose MolecBioNet, a novel graph-based framework that integrates molecular and biomedical knowledge for robust and interpretable DDI prediction. By modeling drug pairs as unified entities, MolecBioNet captures both macro-level biological interactions and micro-level molecular influences, offering a comprehensive perspective on DDIs. The framework extracts local subgraphs from biomedical knowledge graphs and constructs hierarchical interaction graphs from molecular representations, leveraging classical graph neural network methods to learn multi-scale representations of drug pairs. To enhance accuracy and interpretability, MolecBioNet introduces two domain-specific pooling strategies: context-aware subgraph pooling (CASPool), which emphasizes biologically relevant entities, and attention-guided influence pooling (AGIPool), which prioritizes influential molecular substructures. The framework further employs mutual information minimization regularization to enhance information diversity during embedding fusion. Experimental results demonstrate that MolecBioNet outperforms state-of-the-art methods in DDI prediction, while ablation studies and embedding visualizations further validate the advantages of unified drug pair modeling and multi-scale knowledge integration.

Problem

• Predict drug-drug interactions with interpretable graph-based methods
• Integrate molecular and network-level data for mechanistic insights
• Overcome independent drug pair modeling limitations in DDI prediction

Innovation

• Integrates molecular and biomedical knowledge graphs
• Uses context-aware and attention-guided pooling strategies
• Employs mutual information minimization for embedding diversity