ProsperousPlus: A Bioinformatics Workflow for Protease Substrate Cleavage Site Prediction and Model Construction...

Proteolytic cleavage is a fundamental biological process that regulates protein maturation, signaling pathways, immune responses, and disease progression. Accurate identification of protease substrate cleavage sites is critical for understanding protease function and substrate specificity. However, experimental determination of cleavage sites remains labor-intensive and time-consuming, motivating the development of efficient and accurate computational prediction methods. Here, we present a detailed protocol for using ProsperousPlus, a computational platform designed for large-scale prediction of protease-specific substrate cleavage sites and construction of customizable models for these predictions. ProsperousPlus integrates multiple sequence-derived scoring functions with an advanced machine learning framework that automates the construction of optimized predictive models. The framework's workflow includes three major functional modules: (1) prediction of cleavage sites using pre-trained models covering more than 100 proteases, (2) training of custom protease-specific models using user-provided datasets, and (3) deployment of trained models for large-scale prediction tasks. This protocol describes the required hardware and software environment, dataset preparation strategies, model training procedures, and interpretation of prediction results, including probability scores and multiple evaluation metrics such as accuracy and AUC. In addition, model interpretation using Shapley additive explanations (SHAP) enables identification of influential sequence features contributing to cleavage specificity. ProsperousPlus can be used either through a standalone software for large-scale, more frequent analyses or through a user-friendly web server for small-scale, ad hoc use. This protocol provides a practical and reproducible workflow for researchers seeking to analyze protease substrate specificity, construct customized prediction models, and perform high-throughput proteolytic analysis. © 2026 Wiley Periodicals LLC. Basic Protocol: Predicting protease-specific substrate cleavage sites and constructing and deploying protease-specific models using the standalone software version of ProsperousPlus Alternate Protocol: Using the web server version.