Extending coarse-grained modeling methods to improve reliability and yield of designed protein nanocages

Abstract

The King lab designs protein nanocages as vaccine platforms and for other downstream applications. In order to characterize and compare vaccine candidates, we need the nanocage assembly process to be robust and reliable. Our academic and industry partners require uniform production and scalability. However, protein cage assembly is a complex process and often fails for unknown reasons. Unpredictable yield and polydispersity, two of the most common failure modes, may be caused by kinetic factors such as trapping in metastable states or unexpected energy barriers. In the past we have focused on improving interface affinity to make cages more stable, without addressing kinetics or assembly dynamics. Expanding our protein design toolkit to include coarse-grained molecular dynamics (CGMD) and Markov state modeling (MSM) will improve our understanding of the geometric and kinetic factors leading to poor assembly. These insights will guide adjustments to individual design campaigns for the purpose of improving nanocage yield and assembly-product uniformity.