Machine Learning for RNA Therapeutics: From Sequence Design to Clinically Aligned Closed-Loop Engineering

This technical review synthesizes recent advances in machine learning-enabled RNA therapeutics, spanning sequence design, mRNA and self-amplifying RNA engineering, oligonucleotide design, delivery-aware optimization, translational assay strategy, manufacturability, and clinically aligned closed-loop engineering. The review focuses on how design-build-test-learn systems can connect molecular design with potency, innate immune risk, stability, delivery constraints, CMC considerations, and clinical developability. The analysis is based solely on publicly available literature, patents, regulatory documents, conference materials, and the author’s prior public posts. The views expressed are personal and do not represent any current, former, or future employer.