DefPred: In-Silico Tool for Predicting, Scanning, and Designing Defensins

DefPred is a comprehensive computational platform dedicated to the identification, analysis, and design of defensins. Defensins are a major family of antimicrobial peptides (AMPs) that serve as a critical component of the innate immune system in plants, invertebrates, and vertebrates. This tool addresses the challenge of identifying novel defensins from large-scale genomic and proteomic data. Web Server: https://webs.iiitd.edu.in/raghava/defpred/ Citation Kaur D, Patiyal S, Arora C, Singh R, Lodhi G and Raghava GPS (2021) In-Silico Tool for Predicting, Scanning, and Designing Defensins. Front. Immunol. 12:780610. https://doi.org/10.3389/fimmu.2021.780610 About the Research Defensins are small, cysteine-rich cationic peptides that exhibit broad-spectrum activity against bacteria, fungi, and viruses. They are characterized by a unique structural fold stabilized by three or four disulfide bonds. DefPred provides a robust framework to classify these peptides and understand their functional properties. Dataset: The models were developed using a large dataset of 854 experimentally validated defensins and 854 non-defensin peptides (randomly selected from UniProt). Methodology: The platform utilizes several machine learning techniques, including Support Vector Machine (SVM), Random Forest, and Extra Trees, based on various sequence-derived features.