This research introduces a Retrieval-Augmented Generation (RAG) system that translates natural language questions into accurate federated SPARQL queries across bioinformatics knowledge graphs.

• Leverages LLMs to generate structured queries from user questions
• Enhances accuracy by using knowledge graph metadata and schema information
• Incorporates a validation step to detect and correct errors in generated queries
• Deployed as an accessible system at chat.expasy.org

Why it matters: This approach significantly improves access to complex biomedical data by allowing researchers to query federated knowledge graphs using natural language, potentially accelerating scientific discovery and medical research.

LLM-based SPARQL Query Generation from Natural Language over Federated Knowledge Graphs