Computer Simulation by Molecular Dynamics as a Tool for Modelling of Molecular Systems

Abstract

A survey is given of methods for simulation of molecular systems on a computer. The various assumptions, approximations and limitations are discussed and the possibility of making comparisons with experimental quantities is assessed. Finally, a number of practical applications of molecular dynamics simulation techniques in chemistry are reviewed.     

Molecular Dynamics Simulation of Self Diffusion in Microclusters

Abstract

Molecular dynamics simulation is carried out to study the mechanism of self diffusion which is characteristic of solid-like microclusters. A two-dimensional system with the Lennard-Jones potential is employed and the temperatures near the triple point of the two-dimensional bulk system are adopted for the simulation. The results show that: a) microclusters consist of two regions, i.e., solid-like cores and liquid-like surface regions, b) the size dependence of the diffusion coefficient for microclusters is weak, and the value of the solid-like core region is not much different from that of the bulk liquid, c) the activation energy of diffusion for microclusters is twenty to thirty times larger than that for the bulk liquid, d) the diffusion mechanism in the solid-like region involves the collective motion of small domains containing ten to twenty atoms which results in the formation of low density regions, sometimes even vacancy clusters, between them, and atoms in the low density regions change their positions to cause diffusion.     

一种水蛭素类融合蛋白与凝血酶作用的动力学模拟

通过将凝血酶抑制剂水蛭素的C端20肽片段嫁接到血小板结合蛋白AnnexinⅤ上,可以期望获得既具有抗凝血活性,同时又具有导向性的新型工程蛋白质分子.利用计算机辅助分子设计手段模拟了该融合蛋白的分子结构,并对该融合蛋白对凝血酶的抑制作用进行了分子动力学模拟,得到了支持上述想法的结果.     

分子动力学模拟与分子生物力学

生物大分子的微观结构动力学决定其生物学功能,其力学-化学耦合规律是分子生物力学的重点关注方向。分子动力学模拟是耦合生物大分子力学-化学性质微观结构动力学基础的有效手段,其结果可用于预测结构-功能关系、指导实验设计和诠释实验结果。本文简要介绍了分子动力学模拟的方法学特点、基本工作原理及其在分子生物力学中的应用,并展望了未来可能的发展方向和应用前景。     

Molecular Dynamics Simulation Studies of the Density and Temperature Dependence of Self-diffusion in a Cylindrical Micropore

We report Molecular Dynamics calculations of radial density profiles and self-diffusion coefficients of Lennard-Jones fluids in a cylindrical pore of radius 2σ, for a wide range of temperatures and densities. At n _p σ³ = 0.825 the self-diffusion coefficient parallel to the pore walls D _∥*. follows a monotonic (nearly linear) increase with kT/ε and is very similar to that of the bulk self-diffusion coefficient D _b *. At n _p σ³ = 0.4 and kT/ε ≤ 1.0 the curve of D _∥* vs. kT/ε shows a distinct inflection in the region 0.7 ≤ kT/ε ≤ 0.9 and values of D _∥* are much less than D _b * decreasing to near solid state values at very low temperatures. At the highest temperature studied, kT/ε = 2.98, D _∥* is almost inversely proportional to density and in a fairly close agreement with that of D _b *. At KT/ε = 0.49, D _∥* is much smaller than D _b *. The motion of adsorbate particles normal to the walls is also discussed.     

Molecular Dynamics Simulation on the Connection Machine

An algorithm for the efficient calculation of macromolecular force fields on the Connection Machine is described. The full force field is separated into bond interactions and non-bonding interactions. Only the latter are implemented on the Connection Machine, the former, less computationally intensive tasks are performed by an existing, conventional molecular dynamics code on the front end. Parallelization of the evaluation of non-bonding interactions is achieved by the Replicated Systolic Loop algorithm introduced in this paper. The algorithm is a variant of the Systolic Loop scheme often used for the computation of 2-particle forces for the classical N-particle problem.     

Lattice Molecular Dynamics

Abstract

We introduce a new method for simulating fluids in which the particles are constrained to lie on a lattice, but the momenta of the particles are permitted any continuous value. The model includes long-range interactions and is shown to obey the standard macroscopic fluid equations, microscopic time reversal symmetry, and detailed balance.     

Systolic Loop Methods for Molecular Dynamics Simulation,Generalised for Macromolecules

Abstract

Systolic loop programs have been shown to be very efficient for molecular dynamics simulations of liquid systems on distributed memory parallel computers. The original methods address the case where the number of molecules simulated exceeds the number of processors used. Simulations of large flexible molecules often do not meet this condition, requiring the three- and four-body terms used to model chemical bonds within a molecule to be distributed over several processors. This paper discusses how the systolic loop methods may be generalised to accommodate such systems, and describes the implementation of a computer program for simulation of protein dynamics. Performance figures are given for this program running typical simulations on a Meiko Computing Surface using different number of processors.     

Simulation Performance of a Non-Equilibrium Molecular Dynamics Method Using Density Difference as Driving Force

Abstract

The simulation performance of two NEMD algorithms, the constant density difference (DD) and the constant chemical potential difference (CPD) methods, has been compared in fluctuations of molar flux for He and CH₄ permeation across the ZSM-5 membrane. The CPD method and the DD method are found to give almost the same performance; however, the former seems slightly superior to the latter in terms of low fluctuations of molar flux though the former needs more CPU time than the latter. An advantage of the DD method is that it can simulate the mixed-gas permeation through a membrane under the specification of high and low pressures and the composition of feed gas. It is shown that the density profile of permeating gas could provide important information about the relative resistance at the entrance, inside, and exit regions for permeation.     

The Influence of Charge Calculation on Molecular Dynamics Simulation of Adenine in Water

Abstract

Molecular dynamics simulations of an aqueous solution of adenine have been performed using different methods of charge calculation to evaluate the influence of the values of the atomic charges on the dynamical results and to incorporate new information about the interaction between adenine and water. Four sets of partial charges where computed using ab-initio methods. In all cases the hydration properties of adenine were similar. These results support the view that the simulations by molecular dynamics, at least for the regime of infinite dilution, are not sensitive with respect to the different sets of partial charges used. A net hydrophobic behavior of the adenine molecule, on the water was observed.     

Molecular Dynamics Simulation on a Parallel Computer

For the purpose of molecular dynamics simulations of large biopolymers we have built a parallel computer with a systolic loop architecture, based on Transputers as computational units, and have programmed it in occam II. The computational nodes of the computer are linked together in a systolic ring. The program based on this topology for large biopolymers increases its computational throughput nearly linearly with the number of computational nodes. The program developed is closely related to the simulation programs CHARMM and XPLOR, the input files required (force field, protein structure file, coordinates) and output files generated (sets of atomic coordinates representing dynamic trajectories and energies) are compatible with the corresponding files of these programs. Benchmark results of simulations of biopolymers comprising 66, 568, 3 634, 5 797 and 12 637 atoms are compared with XPLOR simulations on conventional computers (Cray, Convex, Vax). These results demonstrate that the software and hardware developed provide extremely cost effective biopolymer simulations. We present also a simulation (equilibrium of X-ray structure) of the complete photosynthetic reaction center of Rhodopseudomonas viridis (12 637 atoms). The simulation accounts for the Coulomb forces exactly, i.e. no cut-off had been assumed.     

Molecular Dynamics Simulation of Collective Motions in Binary Liquids

Collective dynamic properties of different kind of binary liquid mixtures have been investigated by molecular dynamics simulation. The study includes both the longitudinal and the transverse current spectra in simple liquid alloys, 1:1 molten salts and liquid binary mixtures of neutral particles with an ionic-like structure. These systems were chosen as representative of binary liquids with different static structures in order to analyse the effects of structural ordering on the mechanisms of dynamic collective properties. The effect of the mass asymmetry between the two species in the mixture has been also discussed from the results for two different mass ratios for each kind of structure. Two length scales have been considered. On the one hand, the hydrodynamic scale (low wave numbers), where the modes for the partial currents of the two species are characterised by very close frequencies. On the other hand, the molecular scale (higher wave numbers), where the characteristic frequencies for the two species show noticeable differences. Vibrational concentration current modes (optic modes) have been found in neutral mixtures though their influence is rather weak, being the collective dynamic properties of this kind of systems dominated by the mass current modes (acoustic modes). On the contrary, in mixtures of charged particles such as molten salts the contribution of the concentration (charge) currents to the collective dynamics is important and optic modes can be characterised by a well-defined frequency for a wide range of wave numbers. It has been observed that heavy particles have a more relevant role on the mass current correlations whereas light particles play a dominant role on the concentration current correlations. The overall results for the three kinds of liquid mixtures analysed in this paper show that both the longitudinal and transverse current spectra are little dependent on the static structure of the system whereas marked differences are revealed when the particles in the system are either neutral or carry an electric charge.     

Non Equilibrium Molecular Dynamics Simulation of Coupled Heat- and Mass Transport Across a Liquid/Vapor Interface

Abstract

The transport properties of bulk liquid, gas and at the gas/liquid interface were studied for two binary Lennard-Jones/spline mixtures by use of nonequilibrium molecular dynamics. One of the mixtures was an ideal isotope mixture, the other a non-ideal mixture. The simulations gave the thermal conductivity, mutual diffusion coefficient, heat flux, mass flux, and the changes in these quantities across the interface. The local entropy production was expressed in terms of fluxes and thermodynamic forces, and numercial estimates are given. It was shown that the largest contribution to the total entropy production occurs in the vapor phase under the chosen conditions. We expect, however that if the mass flux were larger, the major contribution to the entropy production would come from the liquid phase.     

Molecular Dynamics Simulation of Behaviours of Non-Polar Droplets Merging and Interactions with Hydrophobic Surfaces

This paper presents a molecular dynamics simulation of the behaviours of non-polar droplets merging and also the fluid molecules interacting with a hydrophobic surface. Such behaviours and transport phenomena are popular in general microchannel flow boiling and two-phase flow. The droplets are assumed to be composed of Lennards-Jones type molecules. Periodic boundary conditions are applied in three coordinate directions of a 3-D system, where there exist two liquid droplets and their vapour. The two droplets merge when they come within the prescribed small distance. The merging of two droplets apart from each other at different initial distances is tested and the possible larger （or critical） non-dimensional distance, in which droplets merging can occur, is discussed. The evolution of the merging process is simulated numerically by employing the Molecular Dynamics （MD） method. For interactions with hydrophobic solid wall, a system with fluid confined between two walls is used to study the wetting phenomena of fluid and solid wall. The results are compared with those of hydrophilic wall to show the unique characteristics of hydrophobic interactions by microscopic methods.     

Crystallographic Refinement and Structure-Factor Time-Averaging by Molecular Dynamics in the Absence of a Physical Force Field

Abstract

The application of Molecular-Dynamics simulation in protein-crystallographic structure refinement has become common practice. In this paper, structure optimizations are described where the driving force is derived only from the crystallographic data and not from any physical potential energy function. Under this extreme condition ab initio structure refinement and the application of structure-factor time averaging was investigated using a small 9 atom test system. Success in ab initio refinement, where the starting atomic positions are randomly distributed, depends on the resolution of the crystallographic data used in the optimization. The presence of high resolution data introduces false minima in the X-ray energy profile, enhancing the search problem significantly. On the same system, we also tested the method of time-averaged crystallographically restrained Molecular Dynamics, again in the absence of a physical force field. In this method, the diffraction data is modelled by an ensemble of structures instead of one single structure. In comparison to conventional single-structure refinement, more reflections were required to determine a correct atomic distribution. A time-averaging simulation at 0.2 nm resolution (40 reflections) yielded an incorrect distribution, although a low R-factor was obtained. Simulations at 0.1 nm resolution (248 reflections) gave both low R-factors, 3 to 4%, and correct atomic distributions. The scale factor between the observed and time-averaged calculated structure factor amplitudes appeared to be unstable, when optimized during a time-averaging simulation. Tests of time-averaged restrained simulations with noise added to the observed structure-factor amplitudes, indicated that noise is modelled when no information in the form of constraints or restraints is available to distinguish it from real data.     

Potential Therapeutic Agents on Alzheimer's Disease through Molecular Docking and Molecular Dynamics Simulation Study of Plant-Based Compounds

Globally Alzheimer's disease (AD) is a highly complex, heterogeneous, and multifactorial neurological disease. AD is categorized clinically through a steady loss in memory and progressive decline of cognitive function. So far, there is no effective cure is available for the treatment of AD. Here, we identified Plant-based compounds (PBCs) from seven therapeutic plants through pharmacophore and pharmacokinetics approaches. Subsequently, we retrieved 65 AD associated proteins by Text Mining approach .We observed the interactions between 39 PBCs with 65 AD-associated targets by using molecular docking. Further, we carried out Molecular dynamics simulation analysis to predict the steady binding of top drug-target complexes. The entire MD simulation results analysis was evidence that seven drug-target complexes consistently interacted during the in silico experiment. The top complexes were the target CHLE interacted with 2 PBCs (Pseudojujubogenin and Anahygrine), target VDAC1 interacted with Withanolide R, target THOP1 interacted with Withaolide R, target AOFB interacted with 2 PBCs (Nardostachysin and Viscosalactone B), and target ACHE interacted with the drug (12-Deoxywithastramonolide). These PBCs have stably and flexibly interacted at the protein‘s active site region. Our results suggest that these PBCs and targets are potential therapeutic candidates for molecular development in AD.     

Screening of commercial cyclic peptide as inhibitor NS5 methyltransferase of Dengue virus through Molecular Docking and Molecular Dynamics Simulation

Dengue has become a major global health threat, especially in tropical and subtropical regions. The development of antiviral agent targeting viral replication is really needed at this time. NS5 methyltransferase presents as a novel antiviral target. This enzyme plays an important role in the methylation of 5''-cap mRNA. Inhibition of the NS5 methyltransferase could inhibit dengue virus replication. In this research, two sites of NS5 methyltransferase (S-Adenosyl methionine/SAM binding site and RNA-cap site) were used as targets for inhibition. As much as 300 commercial cyclic peptides were screened to these target sites by means of molecular docking. Analysis of ligand-enzyme binding free energy and pharmacological prediction revealed two best ligands, namely [Tyr123] Prepro Endothelin (110-130), amide, human and Urotensin II, human. According to molecular dynamic simulation, both ligands maintain a stable complex conformation between enzyme and ligand at temperature 310 K and 312 K. Hence, Urotensin II, human is more reactive at 312 K than at 310 K. However, both ligands can be used as potential inhibitor candidates against NS5 methyltransferase of dengue virus with Urotensin II, human exposes more promising activity at 312 K.     

Molecular Dynamics Analysis of Chymosin in Solution and in a Crystalline Environment

The method of molecular dynamics in explicit solvent was applied to test the hypothesis of the existence of a self-inhibited form of chymosin in solution. The paths and energies were calculated for chymosin in solution and in a crystalline environment. The modeling revealed that the intermolecular contacts of chymosin in crystal have negligible influence on the energy stabilization of its self-inhibited conformation. On the other hand, upon molecular dynamics simulation of the active and self-inhibited forms in solution their conformational energies proved to be quite close and the potential barrier between them relatively low. All this supports the possibility of chymosin to adopt spontaneously the self-inhibited conformation in solution, and indicates that it is one of the really existing enzyme forms rather than a crystal packing artifact. The results obtained open novel approaches to studying the specificity of chymosin as well as other aspartic proteinases.     

Molecular Dynamics Studies of Light Hydrocarbons Diffusion in Zeolites

Molecular dynamics simulation techniques have been used to study the diffusion of methane, ethane, propane and i-butane into the zeolite ZSM-5. From the trajectories, the mean-square displacements were obtained and the diffusion coefficients determined using Einstein's diffusion equation. The results, when compared to the available experimental data, indicate that the simulations can provide a realistic representation of the microscopic process of diffusion into the zeolite pores. This revised version was published online in June 2006 with corrections to the Cover Date.     

Modelling Diffusion in Zeolites by Molecular Dynamics Simulations

Abstract

The diffusion of molecules sorbed in zeolites is of growing interest for understanding the mechanisms of chemical processes with regard to selectivity and reactivity [1].

MD simulations give insight into physical systems on the molecular level allowing to study and visualize the motion of molecules even beyond the possibilities of experiments [2,3]. Single system parameters can easily be varied to study their influence, also those parameters that are fixed in reality (e.g., the size of particles). We present a cross section of our recent work to illustrate the capabilities of MD: The self diffusion coefficients (D) of a mixture of methane and xenon in silicalite show remarkable deviations from those of the pure species. This is shown and confirmed by PFG NMR experiments [4].

Simulating ethane in zeolite A the mechanism of diffusion has been studied. The effects of rotation on the diffusion lead to cases where D decreases with growing temperature [5].

The independence of self diffusion on lattice vibrations is proven even for zeolites with windows of guest particle size comparing simulations with rigid and vibrating zeolite lattice [6].