Computer Simulations of Viral Capsid Assembly with Patchy Particle Model

Abstract

Most viruses use the power of self-assembly to protect and transport their genomic material with the help of a protein capsid. Different computational approaches have been employed to understand the dynamics of the assembly pathway, ranging from molecular dynamics simulations with atomistic detail through Brownian dynamics simulations with coarse-grained interactions to statistical or thermodynamic approaches without any explicit representation of space. We have used coarse-grained Brownian dynamics simulations of a patchy particle model to represent the physically correct assembly dynamics in time and space and at the same time to achieve sufficiently long simulation times to study the statistics of complete capsid self-assembly. In particular, we have developed algorithms that account for the anisotropic translational and rotational diffusion of assembly intermediates and ensure detailed balance for reversible patchy particle models (Klein et al., J. Chem. Phys. 140:184112, 2014). We report on studies of two biologically important questions that can be addressed only in such a computer time-efficient framework: the role of event-driven reactivity in capsid self-assembly (Baschek et al., BMC Biophysics 5:22, 2012) and the role of the rate of capsomere addition (Boettcher et al., preprint 2014). We find that hierarchical assembly is more favorable than direct assembly in the case of stable bonds and that slower capsomere supply is more favorable in order to attain a steady state of continuous capsid assembly.