Monte Carlo simulations for NMR-based protein structure determination, aqueous interfacial properties, and pressure-volume curves

Monte Carlo simulations are used in three applications: assessment of residual dipolar couplings (RDCs) obtained from NMR spectroscopic measurements in the prediction of the structure of small helical proteins; structural and chemical properties of the neat aqueous liquid/vapor interface and simple salt solutions; use of the TraPPE (Transferable Potentials for Phase Equilibria) force field to model systems at high temperature and pressure. A Monte Carlo simulated annealing search is developed to optimize the three-dimensional structure of helical protein subunits. A simple residue-based force field is employed to describe the interactions of the interhelical residues. In addition, simulations with orientational constraints, distance constraints, or a combination of both suggest that using only orientational constraints, in the form of RDCs, is inadequate for constraining the search for tertiary structure. However, when combined with limited additional distance data, RDCs can improve the overall RMSD of the model. Configurational-bias Monte Carlo simulations in the Gibbs ensemble are used to investigate the liquid/vapor interface for neat water and water-ion mixtures at 298 K and 1 atm. A mixture of ions of varying size all having the same fixed charge and Lennard-Jones well depth (i.e., the same polarizability) are considered. These simulations identify size effects as having a prominent role in the preference of large inorganic ions for the interface. Simulations in the isobaric-isothermal ensemble at high pressure and temperature compare the united-atom version of the TraPPE force field and single-site exp-6 representations for methane, methanol, oxygen, and ammonia. The TraPPE models, despite being parameterized to the vapor-liquid coexistence curve (i.e. relatively mild conditions), perform remarkably well at extreme conditions. The single-site exp-6 models can fit available experimental data in this region very well, but the parameters are less transferable to conditions below the critical temperature.