DTU Health Tech

Department of Health Technology

ImmunoGeNN - 1.0

input protein sequences and predicts population-level immunogenicty risk scores (MHC-II) based on allele frequencies in the global population

ImmunoGeNN accepts input protein sequences and predicts population-level immunogenicty risk scores (MHC-II) based on allele frequencies in the global population. It also supports deimmunization and immunization of the first input sequence by screening all single amino acid variants (SAVs) for predicted immunogenicity risk changes and sequence likelihood (ESM2). Read more in the EurIPS 2025 SIMBIOCHEM workshop paper

Submit data


Sequence submission

Paste the sequence(s):
or upload a local file:

Options

Mode




Configure reference filtering

Instructions:

Input format

  • Input FASTA - Input FASTA file of protein sequences (minimum sequence length of 15 residues)
  • Human reference filtering - Screens peptide (9mer) binding cores against the human proteome, and assigns pIRS=0 if found.
  • Extra references - Optionally, provide extra reference proteomes in FASTA format to screen against (e.g. known self-proteins)

Modes:

  • Screening mode - Predicts pIRS scores for all input sequences
  • Deimmunization mode - Screens all single amino acid variants (SAVs) in the specified ranges for the first input sequence for predicted pIRS. Generates FASTA and scores.csv file for the top n deimmunizing variants (sorted by ranked pIRS x ranked ESM2 likelihood^2)
  • Immunization mode - Same as above, for variants increasing pIRS.

Output format

ImmunoGeNN predicts per-peptide IRS scores ("pIRS") for the Global population, DRB1 gene class. Sequence pIRS_sum scores are calculated by summing across all peptide pIRS scores in the given sequence.

pIRS score interpretation: We suggest using a pIRS rank threshold of ~83% to identify immunogenic peptides, as described in the paper. Higher scores indicate higher global population (MHC-II presentation) immunogenicity risk, with an estimated experimental Spearman R MAPPs correlation of ~0.45.

Download

ImmunoGeNN 1.0 is freely available on GitHub.

https://github.com/mnielLab/ImmunoGeNN


GETTING HELP

If you need help regarding technical issues (e.g. errors or missing results) contact Technical Support. Please include the name of the service and version (e.g. NetPhos-4.0) and the options you have selected. If the error occurs after the job has started running, please include the JOB ID (the long code that you see while the job is running).

If you have scientific questions (e.g. how the method works or how to interpret results), contact Correspondence.

Correspondence: Technical Support: