UniProt

Provides access to UniProt protein data via MCP for single and batch queries.

python

0

GitHub Stars

python

Language

7 months ago

First Indexed

3 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "mcp-mirror-takumiy235_uniprot-mcp-server": {
      "command": "uv",
      "args": [
        "--directory",
        "path/to/uniprot-mcp-server",
        "run",
        "uniprot-mcp-server"
      ]
    }
  }
}

The UniProt MCP Server lets you fetch protein information directly from UniProt through a Model Context Protocol (MCP) interface. It provides protein names, functions, full sequences, sequence lengths, and organism data, with support for single queries and batch lookups. This makes it easy to power AI assistants with up-to-date protein data and to integrate UniProt data into your workflows.

How to use

You interact with the UniProt MCP Server through an MCP client to request information for one or more UniProt accession numbers. Use the single-protein endpoint to retrieve details for a specific accession, or the batch endpoint to compare multiple proteins at once. The server handles common errors gracefully and caches results to improve performance on repeated requests.

Typical use cases include looking up a protein’s name and function, obtaining the full amino acid sequence, checking the sequence length, or confirming the organism. You can perform batch queries to compare several proteins in a single request and then present a consolidated report to users or downstream systems.

How to install

Prerequisites: you need Python 3.10 or higher installed on your system.

Clone the UniProt MCP Server repository and navigate into it.

git clone https://github.com/TakumiY235/uniprot-mcp-server.git
cd uniprot-mcp-server

Install dependencies using uv (recommended) or pip (alternate).

# Using uv (recommended)
uv pip install -r requirements.txt

# Or using pip
pip install -r requirements.txt

Additional sections

Configuration for Claude Desktop is provided as a JSON snippet. Add a server entry with the MCP command to run the UniProt MCP Server within your Claude Desktop settings. This enables the client to discover and communicate with the MCP server.

{
  "mcpServers": {
    "uniprot": {
      "command": "uv",
      "args": ["--directory", "path/to/uniprot-mcp-server", "run", "uniprot-mcp-server"]
    }
  }
}

Notes on usage and capabilities

The server provides individual and batch protein information, including protein name, function description, full sequence, sequence length, and organism. It caches results for 24 hours to improve performance and includes error handling and logging to aid troubleshooting.

Available tools

get_protein_info

Retrieve information for a single UniProt accession, including name, function, sequence, length, and organism.

get_batch_protein_info

Retrieve information for multiple UniProt accessions in a single call, returning an array of protein information objects.