Overview
MetaboT is an LLM multi-agent framework that turns metabolomics questions into SPARQL queries over knowledge graphs.
Specialized agents perform question checking, entity resolution, schema-aware querying and result interpretation through a CLI and Streamlit application.
Configured LLM providers receive the information included in requests; graph endpoints receive queries. Generated Python uses trusted-mode controls and a subprocess path, which should not be treated as proof of a secure sandbox. Code is Apache-2.0.
Key Features
- Specialized agents perform question checking, entity resolution, schema-aware querying and result interpretation through a CLI and Streamlit application.
Use Cases
- Explore metabolomics knowledge graphs using natural-language questions.
How to Use
Prepare Python 3.11, a supported provider key and the required graph configuration. Use the documented CLI or Streamlit launch procedure and inspect generated queries before sensitive-data use.
See the official repository for installation and interface details: https://github.com/HolobiomicsLab/MetaboT
Sources and verification
This entry describes capabilities documented by the linked project sources. Catalog edits are not evidence of a successful run, compatibility test or scientific performance evaluation.
Read the editorial and verification policy