harinlee83/kinase-library-mcp

Kinase Library MCP Server

A Model Context Protocol (MCP) server for The Kinase Library, providing phosphoproteomics analysis tools through a standardized natural language interface.

https://github.com/user-attachments/assets/f14a8d53-2fef-49c1-b9f6-3fe4f6e85417

Quickstart

Use one of the following to run the server and point your MCP client at it (see MCP Client Integration below):

# Using uv (recommended)
uv run python -m kinase_library_mcp.server

# Or classic Python (after activating a venv)
python -m kinase_library_mcp.server

# Or Docker (after building the image)
docker run -i --rm kinase-library-mcp

Installation

Method 1: Using uv (Recommended)

1. Clone this repository:

git clone https://github.com/harinlee83/kinase-library-mcp.git
cd kinase-library-mcp

2. Install uv if you haven't already:

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

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

# Or with pip
pip install uv

3. Install dependencies:

# Create virtual environment and install dependencies
uv sync --dev

# Or install in editable mode
uv pip install -e ".[dev]"

4. Run the MCP server:

# Using uv (recommended)
uv run python -m kinase_library_mcp.server

# Or activate the environment first
source .venv/bin/activate  # On Windows: .venv\Scripts\activate
python -m kinase_library_mcp.server

Method 2: Classic Python with venv and pip

1. Clone this repository:

git clone https://github.com/harinlee83/kinase-library-mcp.git
cd kinase-library-mcp

2. Create and activate a virtual environment:

# Create virtual environment
python -m venv venv

# Activate virtual environment
source venv/bin/activate  # On Windows: venv\Scripts\activate

3. Install dependencies:

# Upgrade pip
pip install --upgrade pip

# Install the package in development mode
pip install -e ".[dev]"

4. Run the MCP server:

python -m kinase_library_mcp.server

Docker Development

1. Build and run with Docker directly:

# Build the image
docker build -t kinase-library-mcp .

# Run the container (interactive stdio for MCP)
docker run -i --rm kinase-library-mcp

See MCP Client Integration → Option 3 for configuring Docker with your MCP client.

Usage

get_scores Tool

The get_scores tool allows you to score phosphorylation site sequences using The Kinase Library's prediction algorithms.

Input Parameters:

• sequence (required): Phosphorylation site sequence (e.g., "VEPPLs*QETF" where * marks the phosphorylation site)
• pp (optional): Whether to include phosphopriming in the prediction (default: False)
• st_fav (optional): Whether to include serine/threonine favorability (default: True)
• non_canonical (optional): Whether to include non-canonical amino acids (default: False)

MCP Client Integration

Configure your MCP client with one of these options (all use stdio transport):

Option 1: uv (recommended)

{
  "mcpServers": {
    "kinase-library": {
      "command": "uv",
      "args": ["run", "python", "-m", "kinase_library_mcp.server"],
      "cwd": "/path/to/kinase-library-mcp"
    }
  }
}

Option 2: Classic Python (with venv activated)

{
  "mcpServers": {
    "kinase-library": {
      "command": "python",
      "args": ["-m", "kinase_library_mcp.server"],
      "cwd": "/path/to/kinase-library-mcp"
    }
  }
}

Note: Ensure your virtual environment is activated, or set command to your venv’s Python (e.g., /path/to/kinase-library-mcp/venv/bin/python).

Option 3: Docker

{
  "mcpServers": {
    "kinase-library": {
      "command": "docker",
      "args": ["run", "--rm", "-i", "kinase-library-mcp"]
    }
  }
}

Docker notes:

• The -i flag is required for stdio interaction with MCP.
• The --rm flag removes the container after exit.
• The image runs the server and waits for stdio input. All dependencies, including The Kinase Library, are preinstalled.

Example Usage:

{
  "tool": "get_scores",
  "arguments": {
    "sequence": "VEPPLs*QETF",
    "pp": false,
    "st_fav": true,
    "non_canonical": false
  }
}

The Kinase Library

This MCP server integrates with The Kinase Library, a comprehensive Python package for analyzing phosphoproteomics data. The library provides tools for:

• Kinase prediction algorithms
• Enrichment analysis
• Visualization tools
• Insights into kinase activities and signaling pathways

For more information about The Kinase Library, visit: https://github.com/TheKinaseLibrary/kinase-library

Roadmap

Future enhancements may include:

• Additional analysis tools from The Kinase Library