RELATE: Relation Extraction in Biomedical Abstracts with LLMs and Ontology Constraints

2025-09-23

AI Summary

Biomedical knowledge graphs (KGs) suffer from incompleteness, hindering drug discovery and clinical decision support. While large language models (LLMs) can extract relations, their outputs are often non-standardized and lack ontology alignment, impeding direct integration into KGs. To address this, we propose the first ontology-constrained, three-stage relation extraction framework: (1) semantic retrieval via SapBERT-enhanced predicate embeddings; (2) LLM-based contextual re-ranking of candidate relations; and (3) explicit modeling of negated assertions—a novel contribution. Evaluated on ChemProt, our method achieves 52% exact match accuracy and 94% Top-10 recall. Applied to 2,400 HEAL abstracts, it filters 99.6% spurious associations while accurately identifying negated relations. This significantly improves both semantic fidelity and ontology compatibility in KG completion.

Technology Category

• Data Mining & Knowledge Management: Linked Open Data, Knowledge Graphs & KB Completion
• Knowledge Representation and Reasoning: Ontologies
• Natural Language Processing: Information Extraction

Application Category

• Semantics and Knowledge: Scalable techniques for the creation, curation, publication, maintenance, and consumption of large, Web-based, structured, reusable, knowledge graphs and ontologies
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics

Abstract

Biomedical knowledge graphs (KGs) are vital for drug discovery and clinical decision support but remain incomplete. Large language models (LLMs) excel at extracting biomedical relations, yet their outputs lack standardization and alignment with ontologies, limiting KG integration. We introduce RELATE, a three-stage pipeline that maps LLM-extracted relations to standardized ontology predicates using ChemProt and the Biolink Model. The pipeline includes: (1) ontology preprocessing with predicate embeddings, (2) similarity-based retrieval enhanced with SapBERT, and (3) LLM-based reranking with explicit negation handling. This approach transforms relation extraction from free-text outputs to structured, ontology-constrained representations. On the ChemProt benchmark, RELATE achieves 52% exact match and 94% accuracy@10, and in 2,400 HEAL Project abstracts, it effectively rejects irrelevant associations (0.4%) and identifies negated assertions. RELATE captures nuanced biomedical relationships while ensuring quality for KG augmentation. By combining vector search with contextual LLM reasoning, RELATE provides a scalable, semantically accurate framework for converting unstructured biomedical literature into standardized KGs.

Problem

• Extracting biomedical relations from abstracts lacks standardization for knowledge graph integration
• LLM outputs require alignment with ontologies to ensure structured representation
• Converting unstructured biomedical literature into standardized knowledge graphs needs scalable framework

Innovation

• Maps LLM-extracted relations to ontology predicates
• Uses similarity retrieval enhanced with SapBERT embeddings
• Employs LLM-based reranking with negation handling