Biomedical KGs
KG: Knowledge Graph
Overview
Biomedical knowledge graphs connect entities such as genes, diseases, drugs, pathways, and patient findings, often combining literature, databases, and clinical records. Researchers build and refine these graphs with information extraction, ontologies, provenance tracking, and temporal or patient-specific context, then use graph embeddings and neural networks to predict associations or rank candidates. Retrieval-augmented language models and graph-based agents use relevant subgraphs to answer biomedical questions, explain mechanisms, and support tasks such as diagnosis and drug discovery. A central challenge is ensuring that graphs are complete and reliable, and that models use retrieved evidence correctly rather than producing plausible but unsupported conclusions.
Latest papers
71 papers9 papers in the last four weeks, up 13% on the four weeks before. 0.1% of all new papers.