Augmented-Nature-UniProt

This MCP server implements the Unofficial UniProt MCP Server, exposing a suite of 26 bioinformatics tools that leverage UniProt's REST API. It is designed to run as an MCP server that communicates via stdio, enabling AI assistants and MCP clients to perform protein searches, retrieve detailed protein information, access sequences and features, explore pathway and interaction data, and perform batch and advanced searches. Tools are organized into core protein analysis, comparative and evolutionary analysis, structure and function analysis, biological context analysis, batch processing and advanced search, literature and cross-references, and data export utilities. Clients can call individual tools by their names (for example, search_proteins, get_protein_info, search_by_gene, get_protein_sequence, get_protein_features, etc.) and supply the required parameters in JSON format. URI templates provide direct access to UniProt data, enabling seamless integration into workflows.

To use the server from an MCP client, configure the client to connect to the server via the provided command and arguments. For example, when using a Docker-based setup, you may reference the container image as shown in the repository example. The server can be started with npm start during development or launched from a deployed environment using the node entry point path specified in your MCP client configuration. Each tool will accept a JSON payload with its required parameters (e.g., accession, query, organism, format) and return results in the requested format (json, fasta, xml, etc.).

Prerequisites:

• Node.js v16 or higher
• npm (or yarn) available in your environment
• Optional: Docker if you prefer containerized deployment

Setup steps:

1. Clone the repository:

git clone <repository-url>
cd uniprot-server

2. Install dependencies:

npm install

3. Build the project (if a build step is required for your setup):

npm run build

4. Run the server:

npm start

5. Add the server to your MCP client configuration (example):

{
  "mcpServers": {
    "uniprot": {
      "command": "node",
      "args": ["/path/to/uniprot-server/build/index.js"],
      "env": {}
    }
  }
}

Docker (optional):

1. Build the image:

docker build -t uniprot-mcp-server .

2. Run the container:

docker run -i uniprot-mcp-server

Notes and tips:

• The MCP server communicates via stdio; ensure your client configuration points to the correct command and arguments.
• The README indicates a Node.js entry point; adjust the path in your mcp_config to the actual built index.js if you customize the build directory.
• If you expose the server in a container, you may need to mount volumes for build outputs or set environment variables for API keys or rate limits if you extend functionality.
• The available tools include search_proteins, get_protein_info, search_by_gene, get_protein_sequence, get_protein_features, and many others covering biology contexts, batch processing, literature references, and data export formats. Parameters for each tool are defined in the README; validate payloads before sending requests.
• URI templates provide direct access like uniprot://protein/{accession}, uniprot://sequence/{accession}, and uniprot://search/{query} for quick integrations.
• If you encounter UniProt API rate limits, consider implementing request throttling in your client or using cached results where feasible.