B3Pred: A Random-Forest-Based Method for Predicting and Designing Blood–Brain Barrier Penetrating Peptides

Welcome to the official repository for B3Pred, a machine-learning-based method for predicting and designing blood–brain barrier penetrating peptides (B3PPs). This resource is designed to support researchers in CNS drug delivery, peptide therapeutics, and computational biology. Web Server: https://webs.iiitd.edu.in/raghava/b3pred/ Citation Kumar, V., Patiyal, S., Dhall, A., Sharma, N., & Raghava, G. P. S. (2021). B3Pred: A Random-Forest-Based Method for Predicting and Designing Blood–Brain Barrier Penetrating Peptides. Pharmaceutics, 13, 1237. https://doi.org/10.3390/pharmaceutics13081237 About the Method B3Pred is a computational tool for predicting blood–brain barrier penetrating peptides (B3PPs) to facilitate drug delivery into the brain. B3PPs can act both as therapeutics and as drug delivery vehicles for CNS-related diseases such as Alzheimer's disease, Parkinson's disease, and glioblastoma. Models were trained, tested, and evaluated on B3PPs obtained from the B3Pdb database, using over 9000 peptide descriptors and multiple machine learning classifiers. The best-performing random-forest model achieved 85.08% accuracy with an AUROC of 0.93. Data sources integrated include: B3Pdb — blood–brain barrier peptide database CPPsite 2.0 — cell-penetrating peptide repository Swiss-Prot / UniProtKB — for negative dataset generation