Efficient simulation of protein surface adsorption using Dissipative Particle Dynamics with specular chain reflection

Open Access
Author:
Stanik, John Andrew
Graduate Program:
Materials Science and Engineering
Degree:
Master of Science
Document Type:
Master Thesis
Date of Defense:
July 31, 2015
Committee Members:
  • Coray M Colina, Thesis Advisor
Keywords:
  • Simulation
  • Protein
  • Adsorption
  • Dissipative Particle Dynamics
  • DPD
Abstract:
Understanding interactions within complex biological systems is essential to study protein function and transport, and to enable design of biocompatible devices. Studying such complex systems through experiment encounters many challenges, including availability and resolution of experimental data, and control over system parameters to be studied. Computer models are frequently employed to explore such systems. Many biological systems of interest, such as protein surface adsorption, cannot be effectively simulated at the atomistic level. In order to simulate these large systems for the durations required for the desired behavior to evolve, atomistic structure is often represented approximately by “coarse-grain” techniques. Dissipative particle dynamics is one simulation technique which makes large size- and time-scales accessible. Current DPD simulations typically represent two surfaces for adsorption, even when the second surface merely serves to bound the opposite one end of the simulation box. To eliminate the computational demand of such a redundant system, here we use a specular reflecting boundary condition as an alternative. This boundary inverts bead Z-velocity at the box ceiling to bounce them back into the simulation. We identify requirements for a successful reflecting boundary. This boundary is validated by comparison with results of a reference system with a second surface and no reflective boundary. Simulation results including surface adsorption, fluid bead density and temperature are used to confirm the equivalence of the results with both boundary methods. Simulation data are evaluated to assess the adsorption behavior of model protein chains of varying geometry onto simulated surfaces of varying hydrophilicity. It is found that such efficient systems with precise parameter control can prove ideal to evaluate a wide range of surface adsorption behavior which may otherwise be impractical to study in detail.