Agent skill

bio-immunoinformatics-mhc-binding-prediction

Predict peptide-MHC class I and II binding affinity using MHCflurry and NetMHCpan neural network models. Identify potential T-cell epitopes from protein sequences. Use when predicting MHC binding for vaccine design or neoantigen identification.

View SKILL.md on GitHub Repository
Stars 2,009
Forks 275

Install this agent skill to your Project

npx add-skill https://github.com/FreedomIntelligence/OpenClaw-Medical-Skills/tree/main/skills/bio-immunoinformatics-mhc-binding-prediction

SKILL.md

Version Compatibility

Reference examples tested with: MHCflurry 2.1+, pandas 2.2+

Before using code patterns, verify installed versions match. If versions differ:

  • Python: pip show <package> then help(module.function) to check signatures
  • CLI: <tool> --version then <tool> --help to confirm flags

If code throws ImportError, AttributeError, or TypeError, introspect the installed package and adapt the example to match the actual API rather than retrying.

MHC Binding Prediction

"Predict which peptides bind to MHC" → Predict peptide-MHC class I and II binding affinity using neural network models to identify potential T-cell epitopes from protein sequences.

  • Python: mhcflurry.Class1PresentationPredictor().predict() for MHC-I
  • CLI: netMHCpan for alternative MHC-I/II predictions

MHCflurry Setup

Goal: Install MHCflurry and download pre-trained prediction models.

Approach: Install via pip and fetch model weights for class I pan-allele or specific allele predictions.

bash
# Install MHCflurry
pip install mhcflurry

# Download prediction models
mhcflurry-downloads fetch

# Download models for specific alleles
mhcflurry-downloads fetch models_class1_pan

MHCflurry Python API

Goal: Predict peptide-MHC binding affinity and presentation scores for a set of peptides.

Approach: Load the Class1PresentationPredictor and call predict() with peptide-allele pairs to obtain IC50, percentile rank, and presentation scores.

python
from mhcflurry import Class1PresentationPredictor

# Load predictor (includes binding and processing scores)
predictor = Class1PresentationPredictor.load()

# Predict for single allele
result = predictor.predict(
    peptides=['SIINFEKL', 'GILGFVFTL', 'NLVPMVATV'],
    alleles=['HLA-A*02:01', 'HLA-A*02:01', 'HLA-A*02:01']
)

# Result columns:
# - mhcflurry_affinity: Predicted IC50 (nM)
# - mhcflurry_affinity_percentile: Percentile rank
# - mhcflurry_presentation_score: Combined binding + processing

print(result)

Interpret Binding Predictions

Goal: Classify peptide-MHC binding strength from predicted IC50 values.

Approach: Apply standard affinity thresholds (strong <50nM, moderate <500nM, weak <5000nM) to categorize binding.

python
def interpret_binding(ic50_nm):
    '''Interpret MHC binding affinity

    IC50 thresholds (commonly used):
    - <50 nM: Strong binder (high confidence epitope)
    - 50-500 nM: Moderate binder (potential epitope)
    - 500-5000 nM: Weak binder (unlikely epitope)
    - >5000 nM: Non-binder

    Percentile rank (recommended):
    - <0.5%: Strong binder
    - 0.5-2%: Moderate binder
    - >2%: Weak/non-binder
    '''
    if ic50_nm < 50:
        return 'strong'
    elif ic50_nm < 500:
        return 'moderate'
    elif ic50_nm < 5000:
        return 'weak'
    else:
        return 'non-binder'

Batch Prediction

Goal: Predict binding for all peptide-allele combinations in a batch.

Approach: Iterate over peptide-allele pairs, call MHCflurry for each combination, and concatenate results into a single DataFrame.

python
from mhcflurry import Class1PresentationPredictor
import pandas as pd

def predict_binding_batch(peptides, alleles):
    '''Predict binding for multiple peptides and alleles

    Args:
        peptides: List of peptide sequences
        alleles: List of HLA alleles (4-digit format)

    Returns:
        DataFrame with predictions for all combinations
    '''
    predictor = Class1PresentationPredictor.load()

    # Create all combinations
    results = []
    for peptide in peptides:
        for allele in alleles:
            pred = predictor.predict(
                peptides=[peptide],
                alleles=[allele]
            )
            pred['peptide'] = peptide
            pred['allele'] = allele
            results.append(pred)

    return pd.concat(results, ignore_index=True)


# Example usage
peptides = ['SIINFEKL', 'GILGFVFTL', 'NLVPMVATV', 'YMLDLQPETT']
alleles = ['HLA-A*02:01', 'HLA-A*03:01', 'HLA-B*07:02']

predictions = predict_binding_batch(peptides, alleles)
print(predictions[['peptide', 'allele', 'mhcflurry_affinity', 'mhcflurry_affinity_percentile']])

Scan Protein Sequence

Goal: Identify all potential MHC-I epitopes within a protein by scanning overlapping peptide windows.

Approach: Generate all k-mers (8-11aa) from the protein, predict binding for each against target alleles, and retain those below the 2% percentile rank cutoff.

python
def scan_protein_for_epitopes(protein_seq, alleles, peptide_lengths=[8, 9, 10, 11]):
    '''Scan protein for potential MHC epitopes

    MHC-I typically binds 8-11mer peptides
    Most common: 9-mers

    Returns all peptides with predicted binding
    '''
    from mhcflurry import Class1PresentationPredictor

    predictor = Class1PresentationPredictor.load()

    epitopes = []
    for length in peptide_lengths:
        for i in range(len(protein_seq) - length + 1):
            peptide = protein_seq[i:i + length]

            for allele in alleles:
                pred = predictor.predict(peptides=[peptide], alleles=[allele])

                if pred['mhcflurry_affinity_percentile'].values[0] < 2.0:
                    epitopes.append({
                        'peptide': peptide,
                        'position': i + 1,
                        'length': length,
                        'allele': allele,
                        'affinity_nM': pred['mhcflurry_affinity'].values[0],
                        'percentile': pred['mhcflurry_affinity_percentile'].values[0]
                    })

    return pd.DataFrame(epitopes)

MHC Class II Prediction

Goal: Predict MHC class II binding for longer peptides (13-25aa) relevant to CD4+ T-cell responses.

Approach: Query the IEDB NetMHCIIpan API since MHCflurry focuses on class I; submit peptide-allele pairs and parse results.

python
def predict_mhc_ii(peptides, alleles):
    '''Predict MHC class II binding

    MHC-II binds longer peptides (13-25 aa)
    Binding core is ~9aa but flanking regions matter

    Note: MHCflurry focuses on class I
    For class II, use NetMHCIIpan or IEDB tools
    '''
    # NetMHCIIpan via IEDB API
    import requests

    url = 'http://tools-cluster-interface.iedb.org/tools_api/mhcii/'

    results = []
    for peptide in peptides:
        for allele in alleles:
            params = {
                'method': 'netmhciipan_ba',
                'sequence_text': peptide,
                'allele': allele,
                'length': '15'
            }

            response = requests.post(url, data=params)
            # Parse response...

    return results

Common HLA Alleles

Goal: Define population-representative HLA allele sets for broad epitope coverage analysis.

Approach: Use curated lists of the most frequent HLA-A and HLA-B alleles covering ~85% of the Caucasian population.

python
# Most common HLA-A alleles (cover ~85% of population)
COMMON_HLA_A = [
    'HLA-A*02:01',  # ~30% Caucasian
    'HLA-A*01:01',  # ~15%
    'HLA-A*03:01',  # ~13%
    'HLA-A*24:02',  # ~10%
    'HLA-A*11:01',  # ~8%
]

# Most common HLA-B alleles
COMMON_HLA_B = [
    'HLA-B*07:02',
    'HLA-B*08:01',
    'HLA-B*44:02',
    'HLA-B*15:01',
    'HLA-B*35:01',
]

def get_patient_alleles(hla_typing_result):
    '''Parse HLA typing result

    Patients have 2 alleles per locus (one from each parent)
    Format: HLA-A*02:01, HLA-A*24:02
    '''
    # Typically 6 alleles: 2 HLA-A, 2 HLA-B, 2 HLA-C
    return hla_typing_result.split(',')

Related Skills

  • immunoinformatics/neoantigen-prediction - Tumor neoantigen discovery
  • immunoinformatics/epitope-prediction - B-cell epitope prediction
  • clinical-databases/hla-typing - Determine patient HLA type

Recommended Agent Skills

Expand your agent's capabilities with these related and highly-rated skills.

FreedomIntelligence/OpenClaw-Medical-Skills

vcf-annotator

Annotate VCF variants with VEP, ClinVar, gnomAD frequencies, and ancestry-aware context. Generates prioritised variant reports.

2,009 275
Explore
FreedomIntelligence/OpenClaw-Medical-Skills

chemist-analyst

Analyzes events through chemistry lens using molecular structure, reaction mechanisms, thermodynamics, kinetics, and analytical techniques (spectroscopy, chromatography, mass spectrometry). Provides insights on chemical processes, material properties, reaction pathways, synthesis, and analytical methods. Use when: Chemical reactions, material analysis, synthesis planning, process optimization, environmental chemistry. Evaluates: Molecular structure, reaction mechanisms, yield, selectivity, safety, environmental impact.

2,009 275
Explore
FreedomIntelligence/OpenClaw-Medical-Skills

bio-alignment-io

Read, write, and convert multiple sequence alignment files using Biopython Bio.AlignIO. Supports Clustal, PHYLIP, Stockholm, FASTA, Nexus, and other alignment formats for phylogenetics and conservation analysis. Use when reading, writing, or converting alignment file formats.

2,009 275
Explore
FreedomIntelligence/OpenClaw-Medical-Skills

sleep-analyzer

分析睡眠数据、识别睡眠模式、评估睡眠质量，并提供个性化睡眠改善建议。支持与其他健康数据的关联分析。

2,009 275
Explore
FreedomIntelligence/OpenClaw-Medical-Skills

metabolomics-workbench-database

Access NIH Metabolomics Workbench via REST API (4,200+ studies). Query metabolites, RefMet nomenclature, MS/NMR data, m/z searches, study metadata, for metabolomics and biomarker discovery.

2,009 275
Explore
FreedomIntelligence/OpenClaw-Medical-Skills

bio-hi-c-analysis-matrix-operations

Balance, normalize, and transform Hi-C contact matrices using cooler and cooltools. Apply iterative correction (ICE), compute expected values, and generate observed/expected matrices. Use when normalizing or transforming Hi-C matrices.

2,009 275
Explore

Didn't find tool you were looking for?

Be as detailed as possible for better results