Unofficial Ensembl MCP Server

A comprehensive Model Context Protocol (MCP) server that provides access to the Ensembl REST API for genomic data, comparative genomics, and biological annotations.

**Developed by Augmented Nature**

Overview

This server enables seamless access to Ensembl's vast genomic database through a standardized MCP interface. It supports gene lookups, sequence retrieval, variant analysis, comparative genomics, regulatory features, and much more across multiple species.

Features

Gene & Transcript Information

• Gene Lookup: Get detailed gene information by Ensembl ID or gene symbol
• Transcript Analysis: Retrieve all transcripts for a gene with structural details
• Gene Search: Search genes by name, description, or identifier with filtering options

Sequence Data

• Genomic Sequences: Extract DNA sequences for any genomic region or feature
• CDS Sequences: Get coding sequences for specific transcripts
• Sequence Translation: Translate DNA sequences to protein sequences
• Repeat Masking: Support for hard and soft repeat masking

Comparative Genomics

• Homolog Detection: Find orthologous and paralogous genes across species
• Phylogenetic Trees: Generate gene family trees in multiple formats
• Cross-Species Analysis: Compare genes and genomes across different organisms

Variant Data

• Variant Retrieval: Get genetic variants in genomic regions
• Consequence Prediction: Predict variant effects on genes and transcripts
• Population Genetics: Access allele frequencies and population data

Regulatory Features

• Regulatory Elements: Access enhancers, promoters, and TFBS data
• Motif Features: Get transcription factor binding motifs
• Cell Type Context: Filter regulatory features by cell type

Cross-References & Annotations

• External Database Links: Get cross-references to PDB, EMBL, RefSeq, etc.
• Coordinate Mapping: Convert coordinates between genome assemblies
• Ontology Terms: Access GO terms and functional annotations

Species & Assembly Information

• Species Lists: Browse available species and assemblies
• Assembly Statistics: Get genome assembly information and statistics
• Karyotype Data: Access chromosome information and banding patterns

Batch Processing

• Batch Gene Lookup: Process multiple genes simultaneously
• Batch Sequence Fetch: Retrieve sequences for multiple regions efficiently

Installation

# Clone or download the server files
cd ensembl-server

# Install dependencies
npm install

# Build the server
npm run build

Usage with Claude Desktop

Setup Instructions

1. Build the server (if not already done):

npm run build

2. Add to Claude Desktop configuration:

• Open Claude Desktop
• Go to Settings → MCP Servers
• Add a new server with:
• Name: ensembl
• Command: node
• Args: /path/to/ensembl-server/build/index.js

3. Restart Claude Desktop to load the server

Available Tools (25 total)

Gene & Transcript Information

• lookup_gene - Get detailed gene information by stable ID or symbol
• get_transcripts - Get all transcripts for a gene with detailed structure
• search_genes - Search for genes by name, description, or identifier

Sequence Data

• get_sequence - Get DNA sequence for genomic coordinates or gene/transcript ID
• get_cds_sequence - Get coding sequence (CDS) for a transcript
• translate_sequence - Translate DNA sequence to protein sequence

Comparative Genomics

• get_homologs - Find orthologous and paralogous genes across species
• get_gene_tree - Get phylogenetic tree for gene family

Variant Data

• get_variants - Get genetic variants in a genomic region
• get_variant_consequences - Predict consequences of variants on genes and transcripts

Regulatory Features

• get_regulatory_features - Get regulatory elements in genomic region
• get_motif_features - Get transcription factor binding motifs in genomic region

Cross-References & Annotations

• get_xrefs - Get external database cross-references for genes
• map_coordinates - Convert coordinates between genome assemblies

Species & Assembly Information

• list_species - Get list of available species and assemblies
• get_assembly_info - Get genome assembly information and statistics
• get_karyotype - Get chromosome information and karyotype

Batch Processing

• batch_gene_lookup - Look up multiple genes simultaneously
• batch_sequence_fetch - Fetch sequences for multiple regions or features

Example Usage in Claude Desktop

Once connected, you can use natural language to access genomic data:

• "Look up the BRCA2 gene and get its sequence"
• "Find orthologs of TP53 in mouse"
• "Get variants in the region chr17:43044295-43125364"
• "Search for insulin-related genes"
• "Get the assembly information for human genome"
• "Translate this DNA sequence to protein: ATGAAACGC..."

Supported Species

The server supports all species available in Ensembl, including:

• Vertebrates: Human, Mouse, Rat, Zebrafish, etc.
• Plants: Arabidopsis, Rice, Wheat, etc.
• Fungi: Yeast, etc.
• Protists: Various protist species
• Metazoa: Drosophila, C. elegans, etc.

Default species is homo_sapiens when not specified.

Input Formats

Genomic Regions

• chr1:1000000-2000000 - Standard format
• 1:1000000-2000000 - Without 'chr' prefix
• ENSG00000139618 - Feature IDs

Gene/Transcript IDs

• Ensembl IDs: ENSG00000139618, ENST00000380152
• Gene symbols: BRCA2, TP53
• RefSeq IDs: NM_000059

Output Formats

Primary Formats

• JSON: Structured data (default for most tools)
• FASTA: Sequence data
• GFF: Genomic feature format
• VCF: Variant call format

Tree Formats

• JSON: Structured tree data
• Newick: Standard phylogenetic format
• PhyloXML: Rich phylogenetic format

Error Handling

The server provides comprehensive error handling:

• Invalid Parameters: Clear validation messages
• API Errors: Detailed error information from Ensembl
• Network Issues: Timeout and connectivity error handling
• Species Validation: Automatic species name validation

Rate Limiting

The server respects Ensembl's rate limiting guidelines:

• Maximum 15 requests per second
• Appropriate delays between batch operations
• Connection pooling for efficiency

Configuration

Environment Variables

• ENSEMBL_BASE_URL: Override default API base URL
• REQUEST_TIMEOUT: Set custom timeout (default: 30000ms)

Species Configuration

• Default species: homo_sapiens
• Automatic species validation
• Support for all Ensembl divisions

API Coverage

This server provides access to major Ensembl REST API endpoints:

Lookup & Search

• /lookup/id/{id} - Gene/transcript lookup
• /search - Gene search functionality

Sequences

• /sequence/id/{id} - Feature sequences
• /sequence/region/{species}/{region} - Genomic sequences

Comparative Genomics

• /homology/id/{id} - Homology data
• /genetree/id/{id} - Gene trees

Variation

• /variation/region/{species}/{region} - Variant data
• /vep/species/{species}/region - Variant effect prediction

Regulation

• /regulatory/species/{species}/region/{region} - Regulatory features
• /regulatory/species/{species}/microarray/{region} - Motif features

Cross-references

• /xrefs/id/{id} - External database references
• /map/coords/{species}/{assembly}/{region} - Coordinate mapping

Information

• /info/species - Available species
• /info/assembly/{species} - Assembly information

Support

For issues related to:

• Server functionality: Check server logs and error messages
• Ensembl data: Refer to Ensembl documentation
• API usage: See Ensembl REST API guide

Contributing

Contributions are welcome! Please ensure:

• TypeScript compliance
• Comprehensive error handling
• Documentation updates
• Test coverage for new features

Related Tools

This server integrates well with other bioinformatics MCP servers:

• UniProt Server: Protein data integration
• AlphaFold Server: 3D structure predictions
• STRING Server: Protein interaction networks
• PDB Server: Structural biology data

About Augmented Nature

This Ensembl MCP Server is developed by Augmented Nature, a company focused on building AI-powered tools for scientific research and discovery.

Citation

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

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