DOI: 10.1093/bioinformatics/btag604 ISSN: 1367-4811

PEPTiGEN: a tool for mining antimicrobial resistance PEPTides using GENe data of public available repositories

Lisa M Meekes, Francesco Tabaro, Michiel L Bexkens, Dimard E Foudraine, Lennard J M Dekker, Theo M Luider, Nikolaos Strepis, Corné H W Klaassen, Wil H F Goessens

Abstract

Motivation

Detecting antimicrobial resistance (AMR) remains challenging due to the complexity and evolution of resistance mechanisms. Liquid chromatography online coupled to tandem mass spectrometry (LC-MS/MS) offers a promising diagnostic tool. Its success, however, depends on an up-to-date database which can be used to target AMR specific peptides.

Results

We present PEPTiGEN, a computational tool that automatically generates tryptic peptides for any prokaryotic gene and its variants. PEPTiGEN was validated both in silico and in vitro, showing 99% accuracy compared to manually generated tryptic peptides and 98% compared to experimental mass spectrometry data. To demonstrate its potential, we used PEPTiGEN to generate the first AMR peptide database by screening publicly available nucleotide AMR sequences using the Comprehensive Antibiotic Resistance Database (CARD). Together, PEPTiGEN and the AMR peptide database are cornerstones for advancing LC-MS/MS applications in AMR detection and clinical diagnostics.

Availability

The PEPTiGEN code and AMR peptide database are publicly available at github (https://github.com/ftabaro/inspection) and Zenodo (https://doi.org/10.5281/zenodo.21196702).

Supplementary information

Supplementary data are available at Bioinformatics online.

More from our Archive