APRANK is an Antigenic Protein and Peptide Ranker.

APRANK is a bioinformatics pipeline that can:

1. be run to rank candidate antigens and epitopes from a pathogen proteome; or
2. be used with curated antigenicity information to generate and train new models.

To download the files needed to run APRANK as a predictor, go to /aprank and follow the instructions. You will also need to download the corresponding protein and/or peptide models from Dryad.

If you want to train APRANK with new species, or if you want to recreate our models from scratch, go to /model-development. This retraining runs the first half of APRANK to parse data from the organisms, so APRANK is needed.

If you find APRANK useful, please cite:
APRANK: Computational Prioritization of Antigenic Proteins and Peptides From Complete Pathogen Proteomes (2021). Alejandro D Ricci, Mauricio Brunner, Diego Ramoa, Santiago J Carmona, Morten Nielsen, Fernán Agüero. Front Immunol. 12: 702552.
DOI: 10.3389/fimmu.2021.702552. PMID: 34335615. PMCID: PMC8320365.

FASTAs from the organisms used to train APRANK: aprank/model-development/01_inputs/

Antigens from bibligraphy for the organisms used to train APRANK: aprank/model-development/11_antigens/

The same antigens after expanding antigenicity via BLAST and kmer expansion: aprank/model-development/11_antigens/expanded_antigens