Welcome to Immunolyser 2.0

Immunolyser 2.0 is an automated web-based tool developed for immunologists to analyse immunopeptidomics data in a seamless pipeline based on the integration of multiple computational tools for peptide analysis. The pipeline generates a comprehensive report summarising the sequence conservation, clustering of similar peptides in groups, and prediction of binding affinity to various MHC allotypes based on the uploaded peptide data. The pipeline has a user-friendly interface available online and a locally executable version, with no requirement of programming experience. Compared to its 1.0 version, Immunolyser 2.0 has the following updates:

• Expanded functionality allowing analysis of murine immunopeptidomic datasets.
• A novel algorithm, MHC-TP, to predict the MHC class I haplotype in silico where genetic MHC typing is unavailable, ethically restricted, or cost-prohibitive for large-scale datasets.
• An offline and operation system-independent version of Immunolyser to ensure data privacy and enable the analysis of datasets of any size.
• Upgraded Pepscanner module with an interactive visualisation of peptide alignment to source proteins and the option for users to upload a background proteome for source protein analysis.

Please refer to the help page for detailed guidelines on how to use the pipeline and analyse the web-based output reports.

If you find Immunolyser 2.0 helpful, please consider citing us in your study:
Prithvi Raj Munday, Sanjay S.G. Krishna, Joshua Fehring, Nathan P. Croft, Anthony W. Purcell, Chen Li, and Asolina Braun. Immunolyser 2.0: an advanced computational pipeline for comprehensive analysis of immunopeptidomic data. Computational and Structural Biotechnology Journal, 2025. (Accepted)

How to use Immunolyser?

Use the Initialiser module to submit a job. Following input will be required to run the analysis.

• Sample file(s): The sample file needed is a csv file and should have a 'Peptide' column. 'Length' and 'Accession' columns (optional), or other additional columns can be present but will not interfere with analysis. Multiple samples can be uploaded and for every sample, multiple replicate files can be uploaded by selecting multiple files.
• Control file(s) (optional): The control file is a csv file and should have a 'Peptide' column. Peptides present in this file are labelled as 'control' in downloadable results files. Users can also upload multiple control files in case control datasets are present across different files.
• Alleles of interest (optional): MHC-I/II alleles of interest can be chosen for peptide-MHC binding predictions.

Demo and example input

A demo report has been generated to demonstrate the organisation and features of analysis results. It can be accessed by clicking on the Demo tab. To test the pipeline end-to-end, the example files (exported from PEAKs) can be downloaded from here and uploaded in the demo initialiser module to run the analysis. The dataset provided and used to generate the demo report are from the recent immunopeptidomic study by Son, Eric T., et al. In addition, we attached example input files exported from other search engines, including Skyline, ProteinPilot, Spectronaut, and DIA-NN.

References

Following are the tools used in the pipeline to conduct the analysis.

1. Seq2Logo 2.0: A method for construction and visualization of amino acid binding motifs and sequence profiles including sequence weighting, pseudo counts and two-sided representation of amino acid enrichment and depletion. Martin Christen Frolund Thomsen; Morten Nielsen, Nucleic Acids Research 2012; 40 (W1): W281-W287.
2. GibbsCluster 2.0: Unsupervised clustering and alignment of peptide sequences. Andreatta M, Alvarez B, Nielsen M, Nucleic Acids Research (2017) doi: 10.1093/nar/gkx248.
3. NetMHCpan 4.2: Improved prediction of CD8+ epitopes by use of transfer learning and structural features. Jonas Birkelund Nilsson, Jason Greenbaum, Bjoern Peters, and Morten Nielsen, Frontiers in Immunology, August 2025, https://doi.org/10.3389/fimmu.2025.1616113.
4. MHCflurry 2.0: Improved Pan-Allele Prediction of MHC Class I-Presented Peptides by Incorporating Antigen Processing T. O'Donnell, A. Rubinsteyn, U. Laserson, Cell Systems 2020. doi.org/10.1016/j.cels.2020.06.010.
5. MixMHCpred 3.0: Tadros et al. redicting MHC-I ligands across alleles and species: How far can we go?, BioRxiv (2024) .
6. MixMHC2pred: Racle J, Michaux J, Rockinger GA et al. Robust prediction of HLA class II epitopes by deep motif deconvolution of immunopeptidomes, Nat Biotechnol 2019;37:1283-1286.

Feedback

Your feedback will help us to improve. Please send your feedback to Chen.Li@monash.edu.