Speaker: Bonnie Danqing ZHU,
Assistant Professor
Department of Chemical and Biological Engineering, The Hong Kong University of Science and Technology
Abstract: Gene therapy, the delivery of genetic material to the cells of a patient for therapeutic benefit, has been increasingly successful over the past decade. The most successful gene delivery vectors are based on adeno-associated viruses (AAV). Although these natural viruses are safe and non-pathogenic, they present several barriers that limit their efficacy in delivery as they were not evolved by nature for human therapeutic applications. Directed evolution, a strategy involves the iterative genetic diversification of a molecule to create a gene pool and functional selection to isolate variants with optimal properties, has thereby emerged as a powerful approach for re-evolving AAVs of novel and improved functions. In parallel, recent advances in deep sequencing technologies allow millions of sequences to be assayed and used for training machine learning (ML) models for prediction of protein properties. Using the combination of directed evolution and ML-guided design, we have engineered ‘designer’ viral variants with greatly improved packaging, diversity, and primary human brain infection capabilities. The extension of such integration of new technologies will have broad utility in the development of next-generation novel delivery vectors for therapeutic applications of many diseases in the near future.
