Advances in AAV capsid engineering: Integrating rational design, directed evolution and machine learning

Abstract

Adeno-associated virus (AAV) has emerged as a highly promising vector for human gene therapy due to its favorable safety profile, versatility, and ability to transduce a wide range of tissues. However, natural AAV serotypes have shortcomings, including suboptimal transduction efficiency, pre-existing immunity, and a lack of tissue specificity, that hinder their therapeutic potential. To address these challenges, significant efforts are being applied to engineer novel AAV capsids. Rational design leverages structural insights to enhance capsid properties, directed evolution enables unbiased selection of superior variants, and machine learning accelerates discovery by computational analysis of high-throughput screening results to enable predictive algorithms. These strategies have yielded novel capsids with improved transduction efficiency, reduced immunogenicity, and enhanced tissue targeting. Future advances that continue to integrate such multi-disciplinary approaches will further drive the clinical translation of AAV-based therapies.

Keywords: AAV; NGS; directed evolution; gene therapy; high-throughput screening; machine learning; next generation sequencing; rational design.