Abstract
With antimicrobial resistance emerging as a top global health concern, new approaches to prevent and treat infections are urgently needed. Antimicrobial peptides are recognized as a promising alternative to conventional antibiotics. Considering the costly, labor-intensive, and time-consuming nature of wet lab screening, in recent years, proteomics research has increasingly relied on in silico methods to predict physicochemical, structural, and biological properties of candidate peptides. Here, we present the AMPSeek protocol to enable scientists to screen peptides and predict their potential antimicrobial activity, toxicity, and associated three-dimensional structure with a single command. The AMPSeek framework integrates AMPlify, tAMPer, and LocalColabFold, scales efficiently with increasing sample size, and produces an interactive HyperText Markup Language report to facilitate result interpretation. We provide step-by-step instructions for installing and running AMPSeek on a ten-peptide test set, along with two additional datasets (n = 20 and n = 100; mean peptide length ≈121.3) to demonstrate AMPSeek's near-linear scalability. AMPSeek is available on GitHub under the GPLv3 software license agreement: https://github.com/bcgsc/AMPSeek. © 2026 The Author(s). Current Protocols published by Wiley Periodicals LLC. Basic Protocol: Prediction of antimicrobial peptide activity, three-dimensional structure, and toxicity.