Alphafold MCP Server

🚀 AlphaFold MCP Server

A comprehensive Model Context Protocol (MCP) server offering access to the AlphaFold Protein Structure Database via a rich set of tools and resources for protein structure prediction analysis.

🚀 Quick Start

This MCP server enables seamless integration with AlphaFold's vast collection of protein structure predictions. It offers various tools for structure retrieval, confidence analysis, batch processing, and visualization preparation, which is perfect for researchers, bioinformaticians, and structural biologists working with predicted protein structures.

✨ Features

🧬 Core Structure Tools

• Structure Retrieval: Retrieve AlphaFold predictions by UniProt ID.
• Multi - format Downloads: Support for PDB, CIF, BCIF, and JSON formats.
• Availability Checking: Verify the existence of predictions for specific proteins.

🔍 Search & Discovery

• Structure Search: Find proteins by name, gene, or organism.
• Organism Browsing: List all available structures for specific species.
• Coverage Statistics: Obtain comprehensive organism - level statistics.

📊 Confidence & Quality Analysis

• Per - residue Confidence: Get detailed confidence scores for each amino acid.
• Region Analysis: Identify high/low confidence structural regions.
• Quality Validation: Assess the overall prediction reliability.

⚡ Batch Processing

• Bulk Retrieval: Process multiple proteins simultaneously.
• Batch Downloads: Efficiently download multiple structures.
• Parallel Analysis: Conduct confidence analysis for protein sets.

🔬 Comparative Analysis

• Structure Comparison: Perform side - by - side analysis of multiple proteins.
• Similarity Search: Find structurally related proteins.
• Coverage Comparison: Analyze prediction completeness.

🎨 Visualization Integration

• PyMOL Scripts: Use ready - to - use visualization scripts.
• ChimeraX Integration: View confidence - colored structures.
• Custom Export Formats: Enjoy flexible data export options.

📦 Installation

# Clone or create the server directory
npm install

# Build the server
npm run build

💻 Usage Examples

Basic Usage

As MCP Server

Add to your MCP configuration:

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

Direct Usage

# Start the server
npm start

# Or run directly
node build/index.js

Advanced Usage

The server provides a wide range of tools for different scenarios. Here are some usage examples for each tool category:

Core Structure Tools

• get_structure: Retrieve AlphaFold structure prediction for a specific UniProt ID.

{
  "uniprotId": "P04637",
  "format": "json"
}

• download_structure: Download AlphaFold structure file in specified format.
• check_availability: Check if AlphaFold structure prediction is available for a UniProt ID.

Search & Discovery Tools

• search_structures: Search for available AlphaFold structures by protein name or gene.
• list_by_organism: List all available structures for a specific organism.
• get_organism_stats: Get statistics about AlphaFold coverage for an organism.

Confidence & Quality Tools

• get_confidence_scores: Get per - residue confidence scores for a structure prediction.
• analyze_confidence_regions: Analyze confidence score distribution and identify high/low confidence regions.
• get_prediction_metadata: Get metadata about the prediction including version, date, and quality metrics.

Batch Processing Tools

• batch_structure_info: Get structure information for multiple proteins simultaneously.
• batch_download: Download multiple structure files.
• batch_confidence_analysis: Analyze confidence scores for multiple proteins.

Comparative Analysis Tools

• compare_structures: Compare multiple AlphaFold structures for analysis.
• find_similar_structures: Find AlphaFold structures similar to a given protein.

Coverage & Completeness Tools

• get_coverage_info: Get information about sequence coverage in the AlphaFold prediction.
• validate_structure_quality: Validate and assess the overall quality of an AlphaFold prediction.

Export & Integration Tools

• export_for_pymol: Export structure data formatted for PyMOL visualization.
• export_for_chimerax: Export structure data formatted for ChimeraX visualization.
• get_api_status: Check AlphaFold API status and database statistics.

📚 Documentation

Available Resources

Resource Templates

• alphafold://structure/{uniprotId}:
  • MIME Type: application/json
  • Description: Complete AlphaFold structure prediction for a UniProt ID
• alphafold://pdb/{uniprotId}:
  • MIME Type: chemical/x-pdb
  • Description: PDB format structure file for a UniProt ID
• alphafold://confidence/{uniprotId}:
  • MIME Type: application/json
  • Description: Per - residue confidence scores for a structure prediction
• alphafold://summary/{organism}:
  • MIME Type: application/json
  • Description: Summary of all available structures for an organism

Example Workflows

Basic Structure Analysis

// 1. Check if structure is available
await use_mcp_tool("alphafold-server", "check_availability", {
  uniprotId: "P04637",
});

// 2. Get structure metadata
await use_mcp_tool("alphafold-server", "get_prediction_metadata", {
  uniprotId: "P04637",
});

// 3. Analyze confidence scores
await use_mcp_tool("alphafold-server", "get_confidence_scores", {
  uniprotId: "P04637",
  threshold: 70,
});

Comparative Study

// Compare multiple related proteins
await use_mcp_tool("alphafold-server", "compare_structures", {
  uniprotIds: ["P04637", "P53350", "P63151"],
});

// Batch confidence analysis
await use_mcp_tool("alphafold-server", "batch_confidence_analysis", {
  uniprotIds: ["P04637", "P53350", "P63151"],
});

Visualization Preparation

// Export for PyMOL with confidence coloring
await use_mcp_tool("alphafold-server", "export_for_pymol", {
  uniprotId: "P04637",
  includeConfidence: true,
});

// Export for ChimeraX
await use_mcp_tool("alphafold-server", "export_for_chimerax", {
  uniprotId: "P04637",
  includeConfidence: true,
});

Organism - wide Analysis

// Get human protein statistics
await use_mcp_tool("alphafold-server", "get_organism_stats", {
  organism: "Homo sapiens",
});

// List available structures
await use_mcp_tool("alphafold-server", "list_by_organism", {
  organism: "Homo sapiens",
  size: 100,
});

API Reference

The server connects to the AlphaFold API at https://alphafold.ebi.ac.uk/api/ and provides structured access to:

• Structure Predictions: Complete protein structure data
• Confidence Scores: Per - residue reliability metrics
• Metadata: Prediction versions, dates, and quality information
• Cross - references: Links to other databases and resources

🔧 Technical Details

Error Handling

The server includes comprehensive error handling for:

• Invalid UniProt IDs
• Missing structure predictions
• API connectivity issues
• Rate limiting and timeouts
• Malformed requests

Rate Limiting

Please be mindful of API usage:

• Batch operations are limited to reasonable sizes
• Large requests are automatically chunked
• Built - in delays prevent API overload

🤝 Contributing

Contributions are welcome! Please ensure:

• TypeScript type safety
• Comprehensive error handling
• Documentation for new features
• Testing with real AlphaFold data

📄 About

Developed by Augmented Nature - augmentednature.ai

Augmented Nature specializes in creating AI - powered tools and infrastructure for scientific research and data analysis.

📄 License

MIT License - see LICENSE file for details.

📖 Citation

If you use this MCP server in your research, please cite:

• AlphaFold Database: https://alphafold.ebi.ac.uk/
• Model Context Protocol: https://modelcontextprotocol.io/

If you use this project in your research or publications, please cite it as follows:

@misc{alphafoldmcp2025, 
author = {Moudather Chelbi},
title = {AlphaFold MCP Server},
year = {2025},
howpublished = {https://github.com/Augmented-Nature/AlphaFold-MCP-Server},
note = {Accessed: 2025-06-29}
}

🆘 Support

For issues and questions:

• Check the AlphaFold API documentation
• Review error messages for debugging hints
• Ensure UniProt IDs are valid and current

⚠️ Important Note

This server provides a convenient interface to AlphaFold data but does not store or cache structure data. All data is retrieved directly from the official AlphaFold API.