Tool name: chembl_query
Description: This tool translates a natural language question into a SQL query, runs it against the ChEMBL database, and returns the results. It is a one-shot tool, only use for simple data lookup, not for reasoning.
It accepts one parameter:
- prompt_input: the natural language question to answer using ChEMBL data.
Examples: "What are the most potent IC50 compounds against EGFR?",
"Give me approved small molecule drugs for type 2 diabetes with their SMILES",
"What targets has aspirin been tested against?"
The tool returns:
- chembl_results: a data table where each row corresponds to one database record matching the query.
- A summary message describing the number of rows returned, key patterns or values in the data, and whether the 50-row limit was likely hit.
Use this tool when the user asks any question that requires looking up compounds, targets, activities, assays, drug indications, or physicochemical properties from the ChEMBL database. If the summary indicates an empty result or a likely query error, call this tool again with a rephrased or simplified prompt_input.
To use this workflow in KNIME, download it from the below URL and open it in KNIME:
Download WorkflowDeploy, schedule, execute, and monitor your KNIME workflows locally, in the cloud or on-premises – with our brand new NodePit Runner.
Try NodePit Runner!