From Knowledge to Treatment: Large Language Model Assisted Biomedical Concept Representation for Drug Repurposing

2025-10-14

AI Summary

Existing biomedical knowledge graphs (KGs) for drug repurposing overlook mechanistic commonsense knowledge—such as drug–indication incompatibility priors—observed in real-world experimental practice, limiting their effectiveness for rare and complex diseases. To address this, we propose LLaDR, the first framework that injects treatment-related mechanistic knowledge encoded in large language models (LLMs) into KG embedding learning. Specifically, LLaDR leverages LLMs to extract semantic textual representations of KG entities and employs them to guide fine-tuning of KG embeddings, thereby enabling synergistic modeling of structured KG knowledge and unstructured LLM-derived commonsense. Experimental results demonstrate substantial improvements in disease–drug relational reasoning, achieving state-of-the-art performance across multiple benchmark tasks. A case study on Alzheimer’s disease further validates the method’s efficacy and robustness. The source code is publicly available.

Technology Category

• Data Mining & Knowledge Management: Linked Open Data, Knowledge Graphs & KB Completion
• Knowledge Representation and Reasoning: Common-Sense Reasoning
• Machine Learning: Large Multimodal Models (LMMs)

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
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation

Abstract

Drug repurposing plays a critical role in accelerating treatment discovery, especially for complex and rare diseases. Biomedical knowledge graphs (KGs), which encode rich clinical associations, have been widely adopted to support this task. However, existing methods largely overlook common-sense biomedical concept knowledge in real-world labs, such as mechanistic priors indicating that certain drugs are fundamentally incompatible with specific treatments. To address this gap, we propose LLaDR, a Large Language Model-assisted framework for Drug Repurposing, which improves the representation of biomedical concepts within KGs. Specifically, we extract semantically enriched treatment-related textual representations of biomedical entities from large language models (LLMs) and use them to fine-tune knowledge graph embedding (KGE) models. By injecting treatment-relevant knowledge into KGE, LLaDR largely improves the representation of biomedical concepts, enhancing semantic understanding of under-studied or complex indications. Experiments based on benchmarks demonstrate that LLaDR achieves state-of-the-art performance across different scenarios, with case studies on Alzheimer's disease further confirming its robustness and effectiveness. Code is available at https://github.com/xiaomingaaa/LLaDR.

Problem

• Enhancing biomedical concept representation in knowledge graphs
• Incorporating common-sense mechanistic priors for drug compatibility
• Improving drug repurposing for complex and rare diseases

Innovation

• LLM extracts enriched biomedical entity representations
• Fine-tunes knowledge graph embedding models with LLM data
• Injects treatment knowledge to improve concept understanding