An End-to-End Method for AI-Driven Peptide Vaccine Formulation and Translation through Deep Learned Exosome Precision Drug Delivery

Artificial intelligent systems are rapidly integrating and transforming pharmaceutical development approaches by priming facile data analysis for establishing an envisioned pathway. Implementation of these systems with exosomes is a completely untapped area, of which could directly lead into maximized efficacy of cellular targeting through the exosome’s characteristic homing ability. By supplanting exosome surfaces with deep learning constructed receptor targeting ligands, a uniquely tailorable system can be created to unlock the targeted pharmaceutical delivery, pharmacological manipulation, and point-cellular signal pathway probing. In this perspective, exosomes delivering peptide vaccines developed in silico are the exact pharmaceutical demonstration depicting this benefit. As such, we developed ExoGAN, a deep learning algorithm, as a framework methodology generating HLA-A2:01 targeting peptides for T cell targeting by exosomes to be employed in cancer immunotherapy.