Ols MCP Server

🚀 OLS MCP Server

Large Language Models often struggle when working with ontologies, frequently hallucinating terms or their identifiers. This server offers a reliable solution for accessing and querying ontologies, guaranteeing accurate and up - to - date information.

This repository houses a Model Context Protocol (MCP) server that provides access to the Ontology Lookup Service (OLS) API. It enables AI assistants to search for and retrieve ontological terms, concepts, and hierarchies from a variety of biological and medical ontologies.

This server is designed to integrate seamlessly with AI assistants such as Claude Desktop, allowing users to query ontologies using natural language. It supports a wide array of ontologies, including the Gene Ontology (GO), Human Phenotype Ontology (HP), and many others.

Without (top part of image below) and with the mcp server engaged (bottom of the figure below):

🚀 Quick Start

The OLS MCP Server is a reliable solution for accessing and querying ontologies. It can be easily integrated with AI assistants like Claude Desktop, enabling users to interact with ontologies using natural language.

✨ Features

The OLS MCP Server provides the following tools:

• 🔍 Search Terms: Conduct flexible searches for terms across ontologies.
• 📚 Search Ontologies: Discover available ontologies and their metadata.
• ℹ️ Get Ontology Information: Retrieve detailed information about specific ontologies.
• 🎯 Get Term Information: Obtain comprehensive details about specific terms.
• 🌳 Get Term Children: Find direct child terms in ontological hierarchies.
• 👨‍👩‍👧‍👦 Get Term Ancestors: Retrieve parent terms and ancestors.
• 🤖 Find Similar Terms: Discover semantically similar terms using LLM embeddings.

📦 Installation

Prerequisites

• Python 3.12 or higher
• uv package manager

Install uv (if not already installed)

# macOS/Linux
curl -LsSf https://astral.sh/uv/install.sh | sh

# Windows
powershell -c "irm https://astral.sh/uv/install.ps1 | iex"

Install the OLS MCP Server

# Clone the repository
git clone https://github.com/seandavi/ols-mcp-server.git
cd ols-mcp-server

# Install dependencies
uv sync

# Install as a tool (optional)
uv tool install .

💻 Usage Examples

Once configured with Claude Desktop, you can use natural language to interact with the OLS API:

Basic Usage

• Searching for Terms:
  • "Search for terms related to diabetes in the Human Phenotype Ontology"
  • "Find all terms containing 'apoptosis' in the Gene Ontology"
• Getting Ontology Information:
  • "Tell me about the Gene Ontology"
  • "What ontologies are available for chemical compounds?"
• Exploring Term Hierarchies:
  • "Show me the children of the term 'metabolic process' in GO"
  • "What are the ancestor terms for HP:0000118?"
• Finding Similar Terms:
  • "Find terms similar to 'heart development' in the Gene Ontology"

📚 Documentation

Configuration

Claude Desktop Configuration

To use this server with Claude Desktop, add the following configuration to your Claude Desktop config file:

• macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
• Windows: %APPDATA%\Claude\claude_desktop_config.json

{
  "mcpServers": {
    "ols-mcp-server": {
      "command": "uv",
      "args": [
        "tool",
        "run",
        "ols-mcp-server"
      ],
      "env": {}
    }
  }
}

Alternative Configuration (if installed as a tool)

If you installed the server as a uv tool, you can use this simpler configuration:

{
  "mcpServers": {
    "ols-mcp-server": {
      "command": "ols-mcp-server",
      "args": [],
      "env": {}
    }
  }
}

Development Setup

Setting up the Development Environment

1. Clone the repository:

   git clone <repository-url>
   cd ols-mcp-server
   

2. Install dependencies:

   uv sync --extra dev
   

3. Activate the virtual environment:

   source .venv/bin/activate  # macOS/Linux
   # or
   .venv\Scripts\activate     # Windows
   

Code Quality

# Format code
uv run ruff format

# Lint code
uv run ruff check

# Type checking
uv run mypy src/

Adding New Features

1. Add new tools in src/ols_mcp_server/server.py using the @mcp.tool() decorator.
2. Create models in src/ols_mcp_server/models.py for structured responses.
3. Update tests to cover new functionality.
4. Update documentation as needed.

API Documentation

The server interacts with the EBI Ontology Lookup Service API v2. Key endpoints:

• Search: https://www.ebi.ac.uk/ols4/api/search
• Ontologies: https://www.ebi.ac.uk/ols4/api/v2/ontologies
• Terms: https://www.ebi.ac.uk/ols4/api/terms
• Hierarchies: https://www.ebi.ac.uk/ols4/api/v2/ontologies/{ontology}/classes/{term}/children

🔧 Technical Details

The server is designed to work with any ontology available through the EBI Ontology Lookup Service. It supports a wide range of ontologies, including GO, EFO, HP, MONDO, ChEBI, UBERON, and many more.

📄 License

No license information is provided in the original document, so this section is skipped.

Troubleshooting

Common Issues

1. Server not found: Ensure the path to the server is correct in your Claude config.
2. Permission errors: Make sure the server script is executable.
3. Network errors: Check your internet connection and firewall settings.
4. Python version: Ensure you're using Python 3.12 or higher.

Contributing

1. Fork the repository.
2. Create a feature branch: git checkout -b feature-name.
3. Make your changes.
4. Run tests: uv run pytest.
5. Format code: uv run ruff format.
6. Submit a pull request.

Acknowledgments

• EBI Ontology Lookup Service for providing the API.
• FastMCP for the MCP framework.
• The ontology communities for maintaining these valuable resources.