UniProt

The UniProt MCP Server is a bioinformatics platform that connects AI assistants directly to the UniProt protein database through 26 specialized tools.

It handles everything from basic protein searches to complex comparative genomics and structural biology analysis.

Features

• 🔍 Advanced Protein Search – Query by name, keywords, gene symbols, or complex filters
• 🧬 Sequence Analysis – Retrieve and analyze amino acid sequences in multiple formats
• 🔬 Structural Biology Tools – Access 3D structure data, domains, and functional features
• 📊 Comparative Analysis – Side-by-side protein comparison and homolog identification
• 🌳 Evolutionary Studies – Ortholog discovery and phylogenetic relationship analysis
• 🗺️ Pathway Integration – Connect proteins to biological pathways and interaction networks
• 📚 Literature Links – Access publications and cross-references to external databases
• ⚡ Batch Processing – Handle multiple protein accessions efficiently
• 🏷️ Taxonomic Classification – Search by detailed organism classification
• 📤 Multiple Export Formats – Output data in GFF, GenBank, EMBL, XML, and JSON formats

Use Cases

• Drug Discovery Research: Pharmaceutical researchers can quickly identify target proteins, analyze their functional domains, and find homologs across species to understand drug interaction potential and side effects.
• Comparative Genomics Studies: Evolutionary biologists can trace protein evolution across species, identify conserved functional regions, and study how protein families have diverged over time using the ortholog and phylogenetic analysis tools.
• Structural Biology Analysis: Protein engineers can access 3D structure information, analyze functional domains, and identify critical binding sites before planning mutagenesis experiments or protein design projects.
• Systems Biology Research: Researchers studying metabolic pathways can map protein interactions, identify key regulatory proteins, and understand how proteins function within larger biological networks.

How to Use It

1. Clone and Install

git clone <repository-url>
cd uniprot-server
npm install
npm run build

2. Docker Installation (Alternative)

docker build -t uniprot-mcp-server .
docker run -i uniprot-mcp-server

3. Add to your MCP client configuration (Claude Desktop example):

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

Core Analysis Tools

Basic Protein Search

• search_proteins: Search by protein name, keywords, or organism
• Parameters: query (required), organism (optional), size (1-500, default 25), format (json/tsv/fasta/xml)

Detailed Information Retrieval

• get_protein_info: Complete protein data by UniProt accession
• get_protein_sequence: Amino acid sequences in FASTA or JSON
• get_protein_features: Functional domains, active sites, binding regions

Gene-Based Analysis

• search_by_gene: Find proteins by gene name or symbol
• Parameters: gene (required), organism (optional), size (optional)

Advanced Research Tools

Comparative Analysis

• compare_proteins: Side-by-side comparison of multiple proteins
• get_protein_homologs: Find homologous proteins across species
• get_protein_orthologs: Identify orthologous proteins for evolution studies

Structural and Functional Analysis

• get_protein_structure: Access PDB structure references
• get_protein_domains_detailed: InterPro, Pfam, and SMART annotations
• get_protein_variants: Disease-associated mutations and variants

Systems Biology Integration

• get_protein_pathways: KEGG and Reactome pathway connections
• get_protein_interactions: Protein-protein interaction networks
• search_by_function: GO term and functional annotation searches

Specialized Features

Batch Operations

• batch_protein_lookup: Process multiple accessions simultaneously
• advanced_search: Complex queries with multiple filter criteria
• search_by_taxonomy: Detailed taxonomic classification searches

Data Export and Validation

• export_protein_data: Multiple specialized formats (GFF, GenBank, EMBL, XML)
• validate_accession: Verify UniProt accession number validity
• get_external_references: Links to PDB, EMBL, RefSeq, Ensembl databases

Resource Templates

• uniprot://protein/{accession} – Complete protein information
• uniprot://sequence/{accession} – FASTA format sequences
• uniprot://search/{query} – Search results

FAQs

Q: How many proteins can I analyze simultaneously with batch processing?
A: The batch processing tool can handle multiple accessions in a single request. While there’s no hard limit specified, it’s recommended to process 50-100 proteins at a time for optimal performance and to avoid API rate limits.

Q: Can I search for proteins using partial gene names or synonyms?
A: Yes, the search functions support partial matches and gene synonyms. The UniProt database includes alternative gene names and symbols, so searching for “INS” will find insulin-related proteins even if the official gene name is different.

Q: What’s the difference between homologs and orthologs in the comparative analysis tools?
A: Homologs are proteins that share common ancestry regardless of species, while orthologs are homologous proteins in different species that typically retain the same function. Use get_protein_homologs for broader evolutionary relationships and get_protein_orthologs for functional conservation studies.

Q: How current is the UniProt data accessed through this server?
A: The server connects directly to UniProt’s REST API, so you’re accessing the most current version of the database. UniProt updates their database regularly, and this server reflects those updates in real-time.

Q: Can I export protein data for use in other bioinformatics tools?
A: Absolutely. The server supports multiple export formats including GFF for genome browsers, GenBank for sequence analysis tools, EMBL for European databases, and XML for custom parsing. FASTA format is available for sequence-based analysis tools.