Comprehensive Protein Analysis

SkillSearch

Comprehensive protein analysis combining InterProScan domain identification with BLAST similarity search to provide complete functional and evolutionary annotation.

Available today. Use it from your connected AI after setup.

Connect ahel once, and every AI you use reads what you have installed.

Then ask your AI: use the Comprehensive Protein Analysis skill

What this skill tells your AI

The instructions your AI receives, as published by spectrai-initiative/innoclaw in .claude/skills/comprehensive-protein-analysis/SKILL.md and read by ahel’s review.

Usage

1. MCP Server Definition

Use the same BioInfoToolsClient class as defined in the protein-blast-search skill.

2. Comprehensive Protein Analysis Workflow

This workflow combines InterProScan domain analysis with BLAST similarity search to provide a complete functional and evolutionary annotation of a protein sequence.

Workflow Steps:

  1. Validate Input - Check protein sequence format
  2. Run InterProScan - Identify functional domains and GO terms
  3. Run BLAST Search - Find similar sequences and homologs
  4. Integrate Results - Combine domain and homology information for comprehensive annotation

Implementation:

from datetime import timedelta

## Initialize client
client = BioInfoToolsClient(
    "https://scp.intern-ai.org.cn/api/v1/mcp/17/BioInfo-Tools",
    "<your-api-key>"
)

if not await client.connect():
    print("connection failed")
    exit()

## Input: Protein sequence to analyze
protein_sequence = """
MALWMRLLPLLALLALWGPDPAAAFVNQHLCGSHLVEALYLVCGERGFFYTPKTRREAEDLQVGQVELGGGPGAGSLQPLALEGSLQKRGIVEQCCTSICSLYQLENYCN
"""

sequence_id = "INS_HUMAN"

## Step 1, 2 & 3: Run comprehensive analysis (InterProScan + BLAST)
result = await client.session.call_tool(
    "analyze_protein",
    arguments={
        "sequence": protein_sequence.strip(),
        "sequence_id": sequence_id,
        "databases": ["Pfam"],     # InterProScan databases
        "evalue": 1e-5,             # BLAST E-value threshold (more stringent)
        "max_hits": 10              # BLAST max hits
    },
    read_timeout_seconds=timedelta(seconds=1200)  # Allow up to 20 minutes
)

## Step 4: Parse and display comprehensive results
result_data = client.parse_result(result)

print(f"{'='*80}")
print(f"Comprehensive Protein Analysis: {sequence_id}")
print(f"{'='*80}\n")

# InterProScan Results
ips_result = result_data.get("interproscan", {})
if ips_result.get("success"):
    ips_data = ips_result.get("results", {})
    domains = ips_data.get('domains', [])
    go_terms = ips_data.get('go_terms', [])

    print("=== DOMAIN ANALYSIS (InterProScan) ===")
    print(f"Execution time: {ips_result.get('time_seconds', '?')} seconds")
    print(f"Domains found: {len(domains)}")
    print(f"GO annotations: {len(go_terms)}\n")

    if domains:
        print("Functional Domains:")
        for domain in domains:
            print(f"  • {domain.get('name', 'N/A')} ({domain.get('database', 'N/A')})")
            if domain.get('description'):
                print(f"    Description: {domain.get('description')}")
            locations = domain.get('locations', [])
            if locations:
                loc = locations[0]
                print(f"    Position: {loc.get('start')}-{loc.get('end')} aa")
        print()

    if go_terms:
        print("Gene Ontology Annotations:")
        for go in go_terms[:5]:  # Show top 5
            print(f"  • {go.get('id', 'N/A')}: {go.get('name', 'N/A')}")
            print(f"    Category: {go.get('category', 'N/A')}")
        if len(go_terms) > 5:
            print(f"  ... and {len(go_terms) - 5} more")
        print()
else:
    print(f"❌ InterProScan failed: {ips_result.get('error', 'Unknown')}\n")

# BLAST Results
blast_result = result_data.get("blast", {})
if blast_result.get("success"):
    hits = blast_result.get('hits', [])

    print("=== HOMOLOGY SEARCH (BLAST) ===")
    print(f"Execution time: {blast_result.get('time_seconds', '?')} seconds")
    print(f"Similar sequences found: {blast_result.get('total_hits', 0)}")
    print(f"E-value threshold: {1e-5}\n")

    if hits:
        print("Top Homologous Proteins:")
        for i, hit in enumerate(hits[:5], 1):
            print(f"  {i}. {hit['uniprot_id']} - {hit.get('organism', 'N/A')}")
            print(f"     Description: {hit['description']}")
            print(f"     Identity: {hit['identity_percent']:.1f}%, E-value: {hit['evalue']:.2e}")
        if len(hits) > 5:
            print(f"  ... and {len(hits) - 5} more matches")
        print()
    else:
        print("No significant homologs found (E-value threshold may be too stringent)\n")
else:
    print(f"❌ BLAST failed: {blast_result.get('error', 'Unknown')}\n")

# Summary
print("=== FUNCTIONAL SUMMARY ===")
if domains:
    print(f"Protein Family: {domains[0].get('name', 'Unknown')}")
if hits:
    most_similar = hits[0]
    print(f"Most Similar Protein: {most_similar['uniprot_id']} ({most_similar['identity_percent']:.1f}% identity)")
    print(f"Organism: {most_similar.get('organism', 'Unknown')}")
print(f"{'='*80}")

await client.disconnect()

Tool Descriptions

BioInfo-Tools Server:

  • analyze_protein: Comprehensive protein analysis combining InterProScan and BLAST
    • Args:
      • sequence (str): Protein sequence in amino acid single-letter code
      • sequence_id (str, optional): Identifier for the query sequence
      • databases (list, optional): InterProScan databases (default: ["Pfam"])
      • evalue (float, optional): BLAST E-value threshold (default: 0.01)
      • max_hits (int, optional): Maximum BLAST hits (default: 10)
    • Returns:
      • interproscan (dict): InterProScan analysis results
        • success (bool): Whether InterProScan completed
        • results (dict): Domains and GO terms
        • time_seconds (float): Execution time
      • blast (dict): BLAST search results
        • success (bool): Whether BLAST completed
        • hits (list): Similar proteins
        • total_hits (int): Number of matches
        • time_seconds (float): Execution time

Input/Output

Input:

  • sequence: Protein sequence (amino acid single-letter code)
  • sequence_id: Optional identifier for the query
  • databases: List of InterProScan databases to query
  • evalue: BLAST E-value threshold (lower = more stringent)
  • max_hits: Maximum number of BLAST hits to return

Output:

  • InterProScan Results:
    • Functional domains with positions
    • Protein family classifications
    • Gene Ontology annotations
  • BLAST Results:
    • Homologous proteins across species
    • Sequence identity and alignment statistics
    • Evolutionary relationships

Analysis Strategy

This comprehensive approach provides:

  1. Structural Information (InterProScan):

    • Domain architecture and organization
    • Functional motifs and active sites
    • Protein family membership
  2. Evolutionary Context (BLAST):

    • Homologs in other species
    • Sequence conservation patterns
    • Potential orthologs and paralogs
  3. Functional Prediction:

    • Combining domain and homology information
    • GO term annotations for molecular function
    • Biological process involvement

Performance Notes

  • Total execution time: 2-20 minutes depending on sequence length
    • InterProScan: 30 seconds to 15 minutes
    • BLAST: 10-90 seconds
    • Both run sequentially in this workflow
  • Timeout recommendation: Set to at least 1200 seconds (20 minutes)
  • E-value tuning: Use lower E-values (e.g., 1e-10) for highly conserved proteins, higher (e.g., 0.01) for divergent families

Use Cases

  • Complete functional annotation of unknown proteins
  • Validate predicted protein functions
  • Study protein evolution and conservation
  • Identify potential drug targets
  • Annotate proteomes and genome sequences
  • Compare protein function across species

Interpretation Tips

  • High domain coverage + high homology: Well-characterized protein with known function
  • Domains but no homologs: Novel protein with conserved domains, function can be inferred from domains
  • Homologs but no domains: May need more sensitive domain detection or represents a novel fold
  • Neither domains nor homologs: Potentially novel protein, may require experimental characterization

Signals

GitHub stars
391
Forks
28
Last commit
Aug 2026
Advanced
Catalog kind
skill
Gateway key
comprehensive-protein-analysis-spectrai-initiative
Source
github.com/spectrai-initiative/innoclaw