This project implements a Model Context Protocol (MCP)-compliant API server that exposes tools to search the EuPMC, a major repository for proteomics data. It allows AI models (such as Claude or other MCP-compatible LLMs) to interact with proteomics datasets programmatically using structured function calling.
- ✅ MCP Server powered by
FastMCP
- 🔍 EuPMC Search Tool to query datasets by keyword, submission date, popularity, etc.
- 🤖 AI-friendly tools for biomedical and proteomics-related research
- ⚡ Supports both
http
(SSE) andstdio
connection modes - 🛠️ Easily extendable with additional tools
Clone the repo and install dependencies:
git clone https://github.com/PRIDE-Archive/mcp_eupmc_search.git
cd mcp_eupmc_search
poetry install # or pip install -r requirements.txt
Start the MCP server with your preferred connection type (http or stdio):
python -m mcp_eupmc_search --connection_type http --port 9999
Command-line Arguments
Argument | Description | Default |
---|---|---|
--connection_type | Type of connection: http or stdio | http |
--port | Port to run the server (for HTTP mode) | 9999 |
Fetches proteomics datasets from the EuPMC database.
Use this when:
- Searching for proteomics research data
- Mass spectrometry dataset queries
- Biomedical dataset exploration (e.g., cancer-related)
- Finding popular or specific proteomics projects
This server works with any LLM that supports Model Context Protocol, including:
- Anthropic Claude
- Google Gemini
- Open-source MCP clients
- Custom RAG pipelines
+---------------------+ Tool Calls +-----------------------------+
| Claude / Gemini AI | <--------------------> | MCP Europe PMC Server |
+---------------------+ | - fetch_publications() |
| - server_status() |
+-----------------------------+
|
v
+---------------------------+
| EuPMC REST API |
| (https://www.ebi.ac.uk/. |
| europepmc/webservices/ |
| rest/search) |
+---------------------------+
MIT License. See LICENSE for details.