HLA Typing from WES: Why It Matters and How to Do It Right

By Lociven · NeoantigenLab · July 2026

Figure 1. Most common HLA class I alleles by locus in European populations.

HLA Typing Series · Part 1 of 2

Before you can predict which neoantigens a patient's tumor is presenting, you need to know which HLA alleles that patient carries. HLA typing is the necessary first step — and it is one that many researchers either skip, outsource, or misunderstand.

This post explains what HLA typing from WES data involves, which tools are used, what the output means, and where errors commonly occur.

Why HLA typing matters for neoantigen research

MHC class I molecules (HLA-A, HLA-B, HLA-C in humans) are the most polymorphic genes in the human genome — over 30,000 alleles are currently catalogued. Each allele has a unique peptide-binding groove that determines which 8–11 mer peptides can be presented on the cell surface.

A neoantigen prediction that uses the wrong HLA alleles is essentially random. If a patient carries HLA-A*03:01 and you predict binding for HLA-A*02:01, your top candidates may not be presented at all in that patient's tumor.

Every patient must be typed individually. There is no population-level shortcut.

HLA typing from WES: how it works

Dedicated HLA typing tools extract reads from the HLA locus (chromosome 6p21.3) from a standard WES BAM file and perform assembly or alignment against a reference panel of known HLA alleles.

The main tools used in research:

  • OptiType: The most widely used tool for HLA class I typing from WES/RNA-seq. Uses integer linear programming to find the best-fitting allele pair. Reports two-field resolution (e.g., HLA-A*02:01). Fast and well-validated.
  • HLA-HD: More computationally intensive but achieves higher resolution (three-field). Better for rare alleles. Preferred in clinical settings where precision matters.
  • POLYSOLVER: Developed at the Broad Institute, but a standalone tool — not part of GATK. Uses its own Bayesian classifier. Widely used across cancer genomics studies; less actively maintained than OptiType.
  • arcasHLA: Optimized for RNA-seq input. Useful when WES coverage at the HLA locus is low.

Understanding HLA nomenclature

HLA allele names follow a hierarchical format:

HLA-A * 02 : 01 : 01 : 01
       ↑    ↑   ↑     ↑
    Gene Serotype Protein Synonymous DNA

  • First field (02): Serological group — historically defined by antibody reactivity
  • Second field (01): Protein sequence — alleles with different second fields have different amino acids in the binding groove and different peptide preferences
  • Third field (01): Synonymous DNA differences — same protein, different codon usage. Rarely clinically relevant.
  • Fourth field (01): Intronic differences. Almost never clinically relevant.

For neoantigen prediction, two-field resolution (e.g., HLA-A*02:01) is sufficient and is what most prediction tools require. Three-field adds little benefit for binding prediction because the peptide-binding groove is defined by the first two fields.

Each patient has six HLA class I alleles

Humans are diploid, so each person carries two alleles at each of HLA-A, HLA-B, and HLA-C — six alleles total. For neoantigen prediction, you run binding predictions against all six.

Some alleles share overlapping peptide-binding motifs. This means a peptide can be predicted to bind multiple alleles — which is actually beneficial for vaccine design (broader coverage).

HLA-A*02:01 is present in approximately 45% of people with European ancestry, making it the most studied allele in the literature. This creates a research bias — results from HLA-A*02:01-focused studies may not generalize to patients with less common alleles.

HLA loss of heterozygosity: when tumors delete their own HLA

Tumors under immune pressure can delete or silence one HLA haplotype — a phenomenon called HLA loss of heterozygosity (HLA-LOH). This is an active immune escape mechanism: by presenting fewer HLA alleles, the tumor reduces the number of neoantigens on its surface.

HLA-LOH occurs in roughly 40% of non-small cell lung cancers and is associated with worse outcomes in patients treated with checkpoint inhibitors. It also directly undermines neoantigen vaccines — neoantigens predicted to bind the deleted allele will not be presented.

To detect HLA-LOH, tools like LOHHLA (Loss Of Heterozygosity in Human Leukocyte Antigen) compare WES coverage at the HLA locus in tumor vs. normal. If you are running neoantigen prediction for a clinical application, HLA-LOH analysis should be part of the pipeline.

Practical checklist

  1. Run OptiType (or HLA-HD for higher resolution) on the normal tissue WES — not the tumor, which may have HLA mutations or deletions
  2. Confirm you have six alleles (two per gene) — if OptiType returns fewer, there may be coverage issues at the HLA locus
  3. Run LOHHLA on the tumor WES to check for HLA-LOH before finalizing your neoantigen candidate list
  4. Input all six alleles into pVACseq — do not pre-filter by allele frequency

The next post covers clonal vs. subclonal neoantigens — why not all mutations in a tumor are equal targets, and how to prioritize candidates based on tumor evolution.

Tags: HLA typing, OptiType, HLA-HD, neoantigen, WES, HLA-LOH, MHC, tumor immunology, pVACseq

Get the protocol

NeoantigenLab Protocol Pack — HLA Typing Step-by-Step ($29)

The full workflow from WES data to HLA calls: OptiType/HLA-HD commands, resolution levels and when WGS is actually needed, and how to feed the result into pVACseq — fact-checked against current tool docs.

Get the Protocol Pack →

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