Cameron F. Abrams

Enhanced Sampling in Molecular Dynamics Using Metadynamics, Replica-Exchange, and Temperature-Acceleration

We review a selection of methods for performing enhanced sampling in 1 molecular dynamics simulations. We consider methods based on collective variable biasing 2

Large-scale conformational sampling of proteins using temperature-accelerated molecular dynamics

We show how to apply the method of temperature-accelerated molecular dynamics (TAMD) in collective variables [Maragliano L, Vanden-Eijnden E (2006) Chem Phys Lett 426:168–175] to sample the conformational space of multidomain proteins in all-atom, explicitly solvated molecular dynamics simulations. The method allows the system to hyperthermally explore the free-energy surface in a set of collective variables computed at the physical temperature. As collective variables, we pick Cartesian coordinates of centers of contiguous subdomains. The method is applied to the GroEL subunit, a 55-kDa, three-domain protein, and HIV-1 gp120. For GroEL, the method induces in about 40 ns conformational changes that recapitulate the t → r′′ transition and are not observed in unaccelerated molecular dynamics: The apical domain is displaced by 30 Å, with a twist of 90° relative to the equatorial domain, and the root-mean-squared deviation relative to the r′′ conformer is reduced from 13 to 5 Å, representing fairly high predictive capability. For gp120, the method predicts both counterrotation of inner and outer domains and disruption of the so-called bridging sheet. In particular, TAMD on gp120 initially in the CD4-bound conformation visits conformations that deviate by 3.6 Å from the gp120 conformer in complex with antibody F105, again reflecting good predictive capability. TAMD generates plausible all-atom models of the so-far structurally uncharacterized unliganded conformation of HIV-1 gp120, which may prove useful in the development of inhibitors and immunogens. The fictitious temperature employed also gives a rough estimate of 10 kcal/mol for the free-energy barrier between conformers in both cases.

Polymers near Metal Surfaces: Selective Adsorption and Global Conformations

We study the properties of a polycarbonate melt near a nickel surface as a model system for the interaction of polymers with metal surfaces by employing a multiscale modeling approach. For bulk properties, a suitably coarse-grained bead-spring model is simulated by molecular dynamics methods with model parameters directly derived from quantum chemical calculations. The surface interactions are parametrized and incorporated by extensive quantum mechanical density functional calculations using the Car-Parrinello method. We find strong chemisorption of chain ends, resulting in significant modifications of the melt composition when compared to an inert wall.

Energetic ion bombardment of SiO2 surfaces: Molecular dynamics simulations

Numerous profile evolution simulation studies strongly suggest that ions reflecting with glancing angles from etched feature sidewalls are responsible for microtrench formation at the feature bottom. Within these studies such reflections are traditionally assumed specular, where the ion retains all of its incident energy. In this study, we gauge the validity of that assumption by describing the distributions of reflected ion energies, Er, reflected ion angles (polar, θr; azimuthal, φr; and total scatter, αr), obtained via MD simulations of Ar+ bombardment of model SiO2 surfaces. We modeled the physics of the surface atom interactions using an empirical interatomic potential energy function developed by Feuston and Garofalini [J. Chem Phys. 89, 5818 (1988)]. We considered Ar+ ion energies, Ei, of 100 and 200 eV, and incident polar angles, θi, of 0°, 30°, 45°, 60°, 75°, and 85°, measured from the macroscopic surface normal. Each (Ei,θi) combination was used to generate a unique roughened model oxide surface...

Ligand Escape Pathways and (Un)Binding Free Energy Calculations for the Hexameric Insulin-Phenol Complex

Cooperative binding of phenolic species to insulin hexamers is known to stabilize pharmaceutical preparations of the hormone. Phenol dissociation is rapid on hexamer dissolution timescales, and phenol unbinding upon dilution is likely the first step in the conversion of (pharmaceutical) hexameric insulin to the active monomeric form upon injection. However, a clear understanding of the determinants of the rates of phenol unbinding remains obscure, chiefly because residues implicated in phenol exchange as determined by NMR are not all associated with likely unbinding routes suggested by the best-resolved hexamer structures. We apply random acceleration molecular dynamics simulation to identify potential escape routes of phenol from hydrophobic cavities in the hexameric insulin-phenol complex. We find three major pathways, which provide new insights into (un)binding mechanisms for phenol. We identify several residues directly participating in escape events that serve to resolve ambiguities from recent NMR experiments. Reaction coordinates for dissociation of phenol are developed based on these exit pathways. Potentials of mean force along the reaction coordinate for each pathway are resolved using multiple independent steered molecular dynamics simulations with second-order cumulant expansion of Jarzynskis equality. Our results for DeltaF agree reasonably well within the range of known experimental and previous simulation magnitudes of this quantity. Based on structural analysis and energetic barriers for each pathway, we suggest a plausible preferred mechanism of phenolic exchange that differs from previous mechanisms. Several weakly-bound metastable states are also observed for the first time in the phenol dissociation reaction.

Structural Basis for Calmodulin as a Dynamic Calcium Sensor

Calmodulin is a prototypical and versatile Ca(2+) sensor with EF hands as its high-affinity Ca(2+) binding domains. Calmodulin is present in all eukaryotic cells, mediating Ca(2+)-dependent signaling. Upon binding Ca(2+), calmodulin changes its conformation to form complexes with a diverse array of target proteins. Despite a wealth of knowledge on calmodulin, little is known on how target proteins regulate calmodulins ability to bind Ca(2+). Here, we take advantage of two splice variants of SK2 channels, which are activated by Ca(2+)-bound calmodulin but show different sensitivity to Ca(2+) for their activation. Protein crystal structures and other experiments show that, depending on which SK2 splice variant it binds to, calmodulin adopts drastically different conformations with different affinities for Ca(2+) at its C-lobe. Such target protein-induced conformational changes make calmodulin a dynamic Ca(2+) sensor capable of responding to different Ca(2+) concentrations in cellular Ca(2+) signaling.

Collapse dynamics of a polymer chain: Theory and simulation

We present a scaling theory describing the collapse of a homopolymer chain in poor solvent. At time t after the beginning of the collapse, the original Gaussian chain of length N is streamlined to form N/g segments of length R(t), each containing g ~ t monomers. These segments are statistical quantities representing cylinders of length R ~ t1/2 and diameter but structured out of stretched arrays of spherical globules. This prescription incorporates the capillary instability. We compare the time-dependent structure factor derived for our theory with that obtained from ultra-large-scale molecular-dynamics simulation with explicit solvent. This is the first time such a detailed comparison of theoretical and simulation predictions of collapsing chain structure has been attempted. The favorable agreement between the theoretical and computed structure factors supports the picture of the coarse-graining process during polymer collapse.

Concurrent dual-resolution Monte Carlo simulation of liquid methane.

We conduct molecular simulations of liquid methane in a system where molecular resolution fluctuates between atomically explicit and spherically symmetric united atoms. An appropriate dual-resolution canonical ensemble is constructed using (a) effective united atom pair potentials and (b) resolution-control potentials that confine explicit and united atoms chiefly to different slabs in the simulation domain. A Monte Carlo simulation is developed to sample this ensemble. We show that compatibility of the united-atom potentials with the explicit potentials in a concurrent simulation can be tuned by adjusting the width of the interface between the two resolution regions and by direct modification of the united-atom pair potentials. Our results lay the groundwork for treatment of larger atomically specific molecules with similar concurrent multiresolution techniques.

The effect of bond length on the structure of dense bead–spring polymer melts

We explored the effect of the ratio of bond length to excluded volume bead diameter, l0/d0, of simple bead–spring homopolymer chains on the resulting liquid structure of dense, monodisperse polymer melts. We conducted NVT molecular dynamics simulations of both bulk and planar confined polymer liquids. We find that both pressure and surface tension increase with l0/d0. We attribute this sensitivity to the reduction in short range liquid ordering, classically observed for simple fluids, when l0/d0 1. We explain this by examining specific correlations between bond orientations and monomer density. This leads to a better understanding of how intramolecular constraints influence intermolecular structure in molecular liquids.

Atomistic Simulation of Silicon Bombardment by Energetic CF3+: Product Distribution and Energies

We present an analysis of results obtained from molecular dynamics simulations of continuous bombardment of the Si surface with CF3+ ions at normal incidence in the energy (Ei) range of 25–200 eV. Our analysis is aimed at understanding how the distributions in products and their kinetic energies depend on Ei. As Ei increases, the product distribution is shifted toward a lower average molecular weight, and with atomic F and molecular CF becoming the most common product species at the higher incident energies. These findings agree well with recent experimental results. The kinetic energy distributions of the products are sensitive to Ei only in that the high-energy tail of the distribution becomes more prevalent with increasing Ei. Linear cascade theory predictions agree reasonably well with our kinetic energy distributions. Individual species product kinetic energy distributions are much more sensitive to Ei, and primarily reflect that most high-energy products are of low molecular weight. The product kinetic energy as a function of ejection angle is also sensitive to Ei, displaying an increasing maximum value with increasing Ei. These results could potentially help in the guidance and interpretation of molecular beam experiments which seek to detect products in simulated fluorocarbon plasma etching of silicon.