AntiTbPred: Prediction of Antitubercular Peptides From Sequence Information Using Ensemble Classifier and Hybrid Features

Welcome to the official repository for AntiTbPred, a machine learning-based platform developed for predicting antitubercular peptides (AntiTbP) using sequence-derived features and ensemble classification approaches. Web Server: http://webs.iiitd.edu.in/raghava/antitbpred/ Citation Usmani, S. S., Bhalla, S., & Raghava, G. P. S. (2018). Prediction of Antitubercular Peptides From Sequence Information Using Ensemble Classifier and Hybrid Features. Frontiers in Pharmacology, 9, 954. https://doi.org/10.3389/fphar.2018.00954 About the Platform AntiTbPred is a computational platform developed for identifying peptides with antitubercular activity against Mycobacterium tuberculosis. The platform utilizes sequence-based machine learning techniques to distinguish antitubercular peptides from antibacterial and non-antibacterial peptides. Tuberculosis remains one of the major infectious diseases worldwide, particularly due to the rise of multidrug-resistant (MDR), extensively drug-resistant (XDR), and totally drug-resistant (TDR) strains. Antitubercular peptides offer a promising alternative therapeutic strategy because of their selective affinity toward mycobacterial cell walls and reduced toxicity. The study developed predictive models using experimentally validated antitubercular peptides and multiple sequence-derived features including amino acid composition, dipeptide composition, terminal residue composition, and binary profile patterns. Key Features Machine Learning Models Support Vector Machine (SVM) Random Forest (RF) Sequential Minimal Optimization (SMO) Naïve Bayes J48 Decision Tree Prediction Features Amino acid composition (AAC) Dipeptide composition (DPC) Binary profile patterns N-terminal and C-terminal residue composition Hybrid sequence features Prediction Modules Antitubercular peptide prediction Protein sequence scanning Peptide analog generation Ensemble classification system