B3P3-v: Detecting Blood-Brain Barrier Penetrating Peptides from Sequences with Nu-SVC

Effectively transporting therapeutic compounds to the brain poses a considerable obstacle in developing drugs for treating diseases associated with the central nervous system. This challenge mainly arises from the blood-brain barrier (BBB), a natural defense mechanism that limits the access of biotherapeutics to their intended locations within the central nervous system. This barrier obstructs effective treatments for numerous neurological disorders. Peptide therapeutics have emerged as a promising avenue in this field, owing to advancements in peptide chemistry. Compared to small molecules, peptides offer advantages such as high potency, selectivity, low toxicity, and reduced risk of drug interactions. However, the discovery and synthesis of Blood-Brain Barrier Penetrating Peptides (BBBPPs) through experimental methods can be time-consuming and expensive. In this study, we propose a BBBPPs prediction model utilizing the Nu-Support Vector Classifier (with an accuracy of 0.84, an AUCROC of 0.91, and an F1-score of 0.84). The model employs a straightforward physio-chemical feature set derived from the composition, transitions, and distribution of amino acids. The B3P3-v is available at https://github.com/Bhadra-labIB3P3-v