Mutation Clinical Trial Matching MCP
@pickleton89
MCP server to query clinicaltrial.gov to identify mutation specific trials
Overview
What is Mutation Clinical Trial Matching MCP?
Mutation Clinical Trial Matching MCP is an MCP server that enables Claude Desktop to search for clinical trials on clinicaltrials.gov based on specific genetic mutations. It is designed for users who want to query clinical trials using natural language descriptions of mutations.
How to use Mutation Clinical Trial Matching MCP?
Install dependencies with uv pip install -r requirements.txt, configure Claude Desktop with the server’s JSON entry (including paths to the virtual environment and project directory), then start Claude Desktop and ask questions such as "What clinical trials are available for EGFR L858R mutations?" or "Are there any trials for BRAF V600E mutations?".
Key features of Mutation Clinical Trial Matching MCP
- Integrates with Claude Desktop via the Model Context Protocol.
- Queries the clinicaltrials.gov API for mutation-based trial search.
- Returns summarized trial results organized by phase.
- Provides resources describing common mutations (e.g., KRAS G12C).
- Built using the PocketFlow Node pattern for modular execution flows.
- Includes error handling, input validation, and detailed logging.
Use cases of Mutation Clinical Trial Matching MCP
- A researcher asks Claude Desktop for trials related to an EGFR L858R mutation.
- A clinician queries for BRAF V600E mutation trials to identify potential study options.
- A patient or advocate inquires about ALK rearrangement trials.
- A user asks for more information about the KRAS G12C mutation.
FAQ from Mutation Clinical Trial Matching MCP
What runtime and dependencies does the server require?
Python 3.7 or above, plus the packages PocketFlow (≥0.0.1), MCP SDK (≥1.0.0), requests (2.31.0), and python-dotenv (1.1.0). All are installed via uv pip install -r requirements.txt.
How is the server configured in Claude Desktop?
You add a JSON entry to ~/Library/Application Support/Claude/claude_desktop_config.json with the command set to the Python interpreter inside your virtual environment and the args pointing to the clinicaltrials_mcp_server.py file.
Where does trial data come from?
All trial data is retrieved from the public clinicaltrials.gov API. No local trial database is maintained.
What is the current stability of the server?
This is the first phase of development. It works for retrieving trials based on mutations, but bugs remain and further refinements are planned.
How are errors or disconnections handled?
If Claude Desktop disconnects from the MCP server, check the log at ~/Library/Logs/Claude/mcp-server-clinicaltrials-mcp.log, restart Claude Desktop, and verify the server is running correctly.