# Immunopeptidomics

Source: https://onco.cc/terms/immunopeptidomics/  
OnCo record `immunopeptidomics` (Term). Data CC BY-NC 4.0, attribute "Data from OnCo (onco.cc)"; commercial use needs a licence.

## TL;DR

This laboratory approach identifies peptides recovered with immune display molecules. It supplies evidence about peptide presentation in the material that was analyzed.

## Summary

A primary melanoma study used mass spectrometry to identify HLA-associated peptides from native tumor tissue and then evaluated selected mutated peptides with immune assays. This is stronger evidence of natural presentation in the sampled material than a binding prediction alone. The analytical observation still has a defined specimen and workflow. It does not automatically establish the abundance of a ligand on every cancer cell, receptor recognition or a treatment effect. A published protocol isolates MHC-peptide complexes by immunoaffinity capture, dissociates the peptides by acid denaturation, fractionates them by HPLC and identifies them by nano-UPLC tandem mass spectrometry, from cell lines and tissues (Purcell and colleagues, 2019). Capture starts from mild-detergent solubilised lysates (Chong and colleagues, 2018; Sarkizova and colleagues, 2020 immunoprecipitated HLA from soluble lysates), so a lysate is not a surface-only survey, and in multi-allele samples a peptide was assigned to the best-scoring allele or alleles by computation (Sarkizova and colleagues, 2020). Identification depends on the analysis: a deep-learning model trained on more than 300,000 synthetic peptides improved HLA peptide identification up to 7-fold and indicated that 87 per cent of the proposed proteasomally spliced HLA peptides may be incorrect (Wilhelm and colleagues, 2021). A quantitative tissue study isolated peptide-HLA complexes from snap-frozen tissue and used formalin-fixed paraffin-embedded blocks only for histology (Science Translational Medicine, 2022).

## Fields

- Kind: Term
- Last checked: 2026-11-10

## Notes

- Reading guide (OnCo inference). In a mixed specimen a recovered HLA-associated peptide does not reveal its cell of origin or its abundance on living cell surfaces. Non-detection does not show biological absence: it also reflects capture, allele coverage, the searchable sequence set and instrument and software sensitivity, and detection does not show safe tumour recognition. No source read here establishes that formalin-fixed tissue is usable for an assay validated on cells or frozen tissue; the lesson's statement that one protocol proposed an adaptation could not be checked (publisher page unavailable), so that claim is withheld and compatibility must be shown for the named method.
- Fictional worked example. A specimen yields peptide A but not peptide B. A has measured support in that preparation. For B, ask whether its allele was recovered, whether the altered sequence was searchable and whether sensitivity was adequate; 'A detected, B absent everywhere' is stronger than the experiment supports. If B is simply missing from the search database it could not have been identified as configured, so its absence is not a clean biological negative, and correcting the database does not by itself guarantee recovery or prove natural display.
- Reading guide: binding prediction estimates a model output, whereas immunopeptidomics identifies recovered peptides, and total-protein measurements are a different question from HLA-associated fragments (a protein can be abundant without any of its peptides being displayed, an inference); presentation, immune response and efficacy are three separate claims.
- Measurement card in words (a reading guide). Input: which tissue preparation and preservation are compatible, and how much material is consumed. Output and units: peptide identities and unique-sequence counts; instrument intensities are method-dependent and not copies per cell. Identification threshold: the sequence database, the false-discovery rule (spectrum or peptide level) and the supporting spectra. Coverage: which HLA classes and alleles were captured and which allele assignments were measured or inferred. Material and cost: protocol-specific input, preservation, consumption, turnaround and analysis cost, requested before a scarce specimen is used. Validation: recovery, reproducibility and reference-peptide checks for the method. Failure modes: low input, poor recovery, mixed cell populations, database gaps and wrong sequence assignments. Limits: non-detection does not prove absence and detection does not prove safe recognition; there is no universal tissue requirement or detection threshold.

## Sources

- Direct identification of clinically relevant neoepitopes presented on native human melanoma tissue by mass spectrometry.: https://doi.org/10.1038/ncomms13404
- Key Parameters of Tumor Epitope Immunogenicity Revealed Through a Consortium Approach Improve Neoantigen Prediction.: https://doi.org/10.1016/j.cell.2020.09.015
- Purcell et al., Mass spectrometry-based identification of MHC-bound peptides for immunopeptidomics (Nature Protocols 2019): https://doi.org/10.1038/s41596-019-0133-y
- Chong et al., High-throughput and Sensitive Immunopeptidomics Platform Reveals Profound Interferon-gamma-Mediated Remodeling of the Human Leukocyte Antigen Self-Peptidome (Molecular and Cellular Proteomics 2018): https://doi.org/10.1074/mcp.TIR117.000383
- Wilhelm et al., Deep learning boosts sensitivity of mass spectrometry-based immunopeptidomics (Nature Communications 2021): https://doi.org/10.1038/s41467-021-23713-9
- Quantitative immunopeptidomics reveals a tumor stroma-specific target for T cell therapy (Science Translational Medicine 2022): https://doi.org/10.1126/scitranslmed.abo6135
- A large peptidome dataset improves HLA class I epitope prediction across most of the human population (Nature Biotechnology 2020): https://doi.org/10.1038/s41587-019-0322-9

## Connected records

- terms: [Finite tissue and specimen allocation](https://onco.cc/terms/finite-tissue-budget/), [Immune measurements and product evidence](https://onco.cc/terms/immune-product-evidence/), [Immunopeptidome](https://onco.cc/terms/immunopeptidome/), [Neoantigen](https://onco.cc/terms/neoantigen/), [Peptide binding, presentation and immune recognition](https://onco.cc/terms/binding-presentation-recognition/), [Peptide-HLA binding and ligand prediction](https://onco.cc/terms/peptide-hla-binding-prediction/)
- pathways: [Antigen presentation & immune editing](https://onco.cc/pathways/antigen-presentation-immunoediting/)

---
JSON: https://onco.cc/api/v1/entities/immunopeptidomics.json