Rewarding Explainability in Drug Repurposing with Knowledge Graphs

2025-09-02

AI Summary

Drug repurposing urgently requires prediction methods that simultaneously achieve high accuracy and interpretability. To address this, we propose REx, the first reinforcement learning framework for knowledge graph link prediction that incorporates a reward mechanism to guide agents in generating scientifically plausible, interpretable reasoning paths. REx further enhances semantic expressiveness by integrating biomedical ontologies, thereby grounding explanations in biologically meaningful concepts. The method jointly leverages policy gradient optimization, multi-hop path search, and domain-knowledge injection. Evaluated on three standard biomedical knowledge graph benchmarks, REx achieves significant improvements in link prediction accuracy while generating explanatory paths that meaningfully recapitulate established biological mechanisms—such as target–pathway–disease associations—thereby substantially improving the trustworthiness and verifiability of AI-driven scientific discovery.

Technology Category

• Knowledge Representation and Reasoning: Ontologies
• Data Mining & Knowledge Management: Linked Open Data, Knowledge Graphs & KB Completion
• Reasoning under Uncertainty: Other Foundations of Reasoning under Uncertainty

Application Category

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

Abstract

Knowledge graphs (KGs) are powerful tools for modelling complex, multi-relational data and supporting hypothesis generation, particularly in applications like drug repurposing. However, for predictive methods to gain acceptance as credible scientific tools, they must ensure not only accuracy but also the capacity to offer meaningful scientific explanations. This paper presents a novel approach REx, for generating scientific explanations based in link prediction in knowledge graphs. It employs reward and policy mechanisms that consider desirable properties of scientific explanation to guide a reinforcement learning agent in the identification of explanatory paths within a KG. The approach further enriches explanatory paths with domain-specific ontologies, ensuring that the explanations are both insightful and grounded in established biomedical knowledge. We evaluate our approach in drug repurposing using three popular knowledge graph benchmarks. The results clearly demonstrate its ability to generate explanations that validate predictive insights against biomedical knowledge and that outperform the state-of-the-art approaches in predictive performance, establishing REx as a relevant contribution to advance AI-driven scientific discovery.

Problem

• Generating scientific explanations for drug repurposing predictions
• Improving credibility of knowledge graph link prediction methods
• Enhancing explanatory paths with biomedical domain ontologies

Innovation

• Reinforcement learning agent identifies explanatory paths
• Reward mechanisms guide scientific explanation properties
• Domain ontologies enrich paths with biomedical knowledge