pVACseq, Part 2: What pVACseq Does and When to Use It

pVACseq · neoantigen prediction · NetMHCpan · MHCflurry · cancer vaccine pipeline

pVACseq Series · Part 2 of 2

What pVACseq Does and When to Use It

pVACseq is the most widely used pipeline for neoantigen prediction from somatic variants. It takes a VCF file and HLA alleles, generates peptide sequences around each mutation, runs binding predictions against multiple algorithms in parallel, and produces a ranked list of candidate neoepitopes.

It is the right tool when you have a VCF from tumor-normal WES and want a systematically generated, reproducible neoantigen list for research or vaccine design. It is not a one-click solution — the output requires biological interpretation and filtering before it is useful.

Figure 1. Left: binding prediction counts by peptide length for HLA-A*02:01. Right: the prioritization funnel from ~2,800 raw predictions to ~20 final vaccine candidates.

Installation and Input Requirements

# Install via pip (Python 3.8+)
pip install pvactools

# Required inputs:
# 1. Annotated VCF (run through VEP first)
vep --input_file tumor_somatic.vcf \
    --output_file tumor_somatic.vep.vcf \
    --format vcf \
    --vcf \
    --symbol \
    --terms SO \
    --tsl \
    --hgvs \
    --fasta GRCh38.fa \
    --offline

# 2. HLA alleles from OptiType or HLA-HD (see Post 17)
# Format: HLA-A*02:01,HLA-A*24:02,HLA-B*07:02,...

Running pVACseq

pvacseq run \
    tumor_somatic.vep.vcf \            # VEP-annotated VCF
    sample_01 \                        # sample name (output prefix)
    "HLA-A*02:01,HLA-A*24:02,HLA-B*07:02,HLA-B*35:01,HLA-C*07:01,HLA-C*04:01" \
    NetMHCpan MHCflurry \              # prediction algorithms (can combine)
    output_dir/ \                      # output directory
    --iedb-install-directory /opt/iedb/ \
    --class-i-epitope-length 8,9,10,11 \  # peptide lengths (pVACtools 2.0+: -e for class I was split into -e1/-e2)
    --binding-threshold 500 \          # IC50 cutoff (nM)
    --percentile-threshold 2 \         # alternatively: top 2% binders
    --minimum-fold-change 1 \          # MT/WT binding ratio filter
    --normal-cov 40 \                  # minimum normal coverage
    --tdna-cov 25 \                    # minimum tumor DNA coverage
    --trna-cov 10 \                    # minimum tumor RNA coverage
    --expn-val 1 \                     # minimum expression (TPM)
    --net-chop-method cterm \          # proteasomal cleavage prediction
    --netmhc-stab \                    # HLA-peptide stability (add Kd½)
    -t 8

Understanding the Output

pVACseq produces several output files. The most important:

  • sample_01.filtered.condensed.ranked.tsv — Final ranked list, one row per candidate neoepitope
  • sample_01.all_epitopes.tsv — All predictions before filtering
import pandas as pd

results = pd.read_csv("output_dir/MHC_Class_I/sample_01.filtered.condensed.ranked.tsv", sep="\t")

# Key columns to check:
# Mutation - gene + amino acid change
# MT Epitope Seq - mutant peptide sequence
# Best MT IC50 Score - lowest predicted binding affinity (nM)
# Best WT IC50 Score - wild-type peptide binding (for fold-change)
# Fold Change - MT/WT ratio (higher = more tumor-specific)
# Tumor DNA VAF - variant allele frequency (clonality proxy)
# Gene Expression - RNA-seq expression level

# Recommended final filters for vaccine design:
high_priority = results[
    (results["Best MT IC50 Score"] < 50) &       # strong binder
    (results["Fold Change"] > 10) &               # tumor-specific (not self)
    (results["Tumor DNA VAF"] > 0.2) &            # clonal
    (results["Gene Expression"] > 2)              # expressed
].sort_values("Best MT IC50 Score")

print(f"Final candidates: {len(high_priority)}")}

Common Mistakes

  • Not filtering by VAF: Subclonal mutations (VAF <0.1) produce neoantigens in only a fraction of tumor cells. Vaccines targeting subclonal mutations have low probability of impacting disease.
  • Ignoring fold change: A mutation in a gene that is also expressed as wild-type can generate peptides that are similar to self. High MT/WT fold change (>10) indicates the mutant peptide is genuinely foreign.
  • Using only one prediction algorithm: Ensemble predictions (NetMHCpan + MHCflurry) are more reliable than either alone. pVACseq handles this natively — always specify at least two algorithms.

Working with spatial transcriptomics and want to map immune infiltration patterns? See the companion blog SpatiaBio — Squidpy tutorials and spatial omics workflows.

댓글

댓글 쓰기

이 블로그의 인기 게시물

What Is a Neoantigen? A Practical Guide for Experimental Researchers

By Lociven · NeoantigenLab · July 2026 If you work in cancer immunology, you have heard the word neoantigen more times than you can count. It appears in grant proposals, immunotherapy papers, and tumor board discussions. But the explanations are usually either too shallow ("mutated peptides the immune system can recognize") or buried in computational methods papers that assume you already know what a VCF file is. This post is for the wet-lab side of the room — the researchers who understand tumor biology and immunology but want a clear picture of what neoantigens are, where they come from, and why the field cares so much about them. Figure 1. Most somatic mutations do not become immunogenic neoantigens. Each step — coding region, MHC binding, processing, T cell recognition — acts as a filter. The core idea A neoantigen is a peptide fragment, derived from a tumor-specific somatic mutation, that can be presented on the surface of a cancer cell via MHC molecule...
자세한 내용 보기

NetMHCpan Explained: How MHC Binding Prediction Works and How to Read the Output

By Lociven · NeoantigenLab · July 2026 You run NetMHCpan on your mutant peptide list and get back a table full of percentile ranks, EL scores, and BA scores. One peptide shows 0.48% rank. Another shows 2.1%. What does that mean, and which one should you follow up on? NetMHCpan is the most widely used MHC-I binding prediction tool in neoantigen research. Most papers cite it, most pipelines include it, and most wet-lab researchers who encounter its output don't fully understand what it's predicting. This post explains the model, the output, and where interpretation goes wrong. What NetMHCpan predicts NetMHCpan predicts the binding affinity between a peptide and an MHC class I molecule. More precisely, it predicts two things: Binding affinity (BA) : Estimated IC50 in nM — the concentration of peptide needed to displace 50% of a reference ligand from the HLA molecule. Lower IC50 = stronger binder. The common threshold is IC50 < 500 nM for "weak binder,...
자세한 내용 보기

Tumor Mutational Burden (TMB): What It Measures, What It Doesn't, and How to Interpret It

By Lociven · NeoantigenLab · July 2026 Your pathology report comes back with a TMB of 14 mut/Mb. The oncologist asks whether this patient is likely to respond to pembrolizumab. What do you actually say? Tumor mutational burden has become one of the most frequently cited biomarkers in solid tumor immunotherapy — and one of the most frequently misunderstood. This post explains what TMB measures, how it is calculated, what the FDA approval actually covers, and what it cannot tell you. What TMB measures — and what it doesn't TMB is a count of somatic mutations per megabase of coding sequence sequenced. That is all it is. It does not measure: Whether any of those mutations produce peptides that bind HLA Whether those peptides are actually presented on tumor cell surfaces Whether the patient has T cells that recognize them Whether the tumor microenvironment is immunologically accessible TMB is a proxy. More mutations → more potential neoantigens → higher p...
자세한 내용 보기