Molecular Dynamics Simulation Time Research Articles

Molecular Dynamics Simulation of Dislocation Plasticity Mechanism of Nanoscale Ductile Materials in the Cold Gas Dynamic Spray Process

The dislocation plasticity of ductile materials in a dynamic process of cold gas spraying is a relatively new research topic. This paper offers an insight into the microstructure and dislocation mechanism of the coating using simulations of molecular dynamics (MD) because of the short MD simulation time scales. The nano-scale deposition of ductile materials onto a deformable copper substrate has been investigated in accordance with the material combination and impact velocities in the particle/substrate interfacial region. To examine the jetting mechanisms in a range of process parameters, rigorous analyses of the developments in pressure, temperature, dislocation plasticity, and microstructure are investigated. The pressure wave propagation’s critical function was identified by the molecular dynamics’ simulations in particle jet initiation, i.e., exterior material flow to the periphery of the particle and substrate interface. The initiation of jet occurs at the point of shock waves interact with the particle/substrate periphery and leads to localization of the metal softening in this region. In particular, our findings indicate that the initial particle velocity significantly influences the interactions between the material particles and the substrate surface, yielding various atomic strain and temperature distribution, processes of microstructure evolution, and the development of dislocation density in the particle/substrate interfacial zone for particles with various impact velocities. The dislocation density in the particle/substrate interface area is observed to grow much more quickly during the impact phase of Ni and Cu particles and the evolution of the microstructure for particles at varying initial impact velocities is very different.

Predicting Hydrophobicity by Learning Spatiotemporal Features of Interfacial Water Structure: Combining Molecular Dynamics Simulations with Convolutional Neural Networks.

The hydrophobicity of functionalized interfaces can be quantified by the structure and dynamics of water molecules using molecular dynamics (MD) simulations, but existing methods to quantify interfacial hydrophobicity are computationally expensive. In this work, we develop a new machine learning approach that leverages convolutional neural networks (CNNs) to predict the hydration free energy (HFE) as a measure of interfacial hydrophobicity based on water positions sampled from MD simulations. We construct a set of idealized self-assembled monolayers (SAMs) with varying surface polarities and calculate their HFEs using indirect umbrella sampling calculations (INDUS). Using the INDUS-calculated HFEs as labels and physically informed representations of interfacial water density from MD simulations as input, we train and evaluate a series of neural networks to predict SAM HFEs. By systematically varying model hyperparameters, we demonstrate that a 3D CNN trained to analyze both spatial and temporal correlations between interfacial water molecule positions leads to HFE predictions that require an order of magnitude less MD simulation time than INDUS. We showcase the power of this model to explore a large design space by predicting HFEs for a set of 71 chemically heterogeneous SAMs with varying patterns and mole fractions.

Revealing the ultra-low-temperature relaxation peak in a model metallic glass

By systematically investigating the relaxation behavior of a model metallic glass based on the extensive molecular dynamics (MD) simulations combined with the dynamic mechanical spectroscopy method, a pronounced ultra-low temperature peak on the loss modulus spectrum was discovered for the first time in MD simulations. It was found that the relaxation peak occurs at a much lower temperature than the typical temperature for the conventional β relation peak as reported in the literature. According to the atomic displacement analysis, we unravel that the reversible atomic motions, rather than the thermal vibrations or local structural rearrangements, mainly contribute to this relaxation peak. We further identify the atomic level mechanism of this fast relaxation process by characterizing the local geometrical anisotropy. Furthermore, by tracing the dynamic behaviors of these “reversible” atoms, we demonstrate the intrinsic hierarchy of the relaxation modes, which are triggered by atomic vibrations and gradually developed to include the reversible and irreversible atomic movements. Our findings provide an important piece of the puzzle about the holistic picture of the rich relaxation processes in metallic glasses.

Variance of Atomic Coordinates as a Dynamical Metric to Distinguish Proteins and Protein-Protein Interactions in Molecular Dynamics Simulations.

Protein dynamics is a manifestation of the complex trajectories of these biomolecules on a multidimensional rugged potential energy surface (PES) driven by thermal energy. At present, computational methods such as atomistic molecular dynamics (MD) simulations can describe thermal protein conformational changes in fully solvated environments over millisecond timescales. Despite these advances, a quantitative assessment of protein dynamics remains a complicated topic, intricately linked to issues such as sampling convergence and the identification of appropriate reaction coordinates/structural features to describe protein conformational states and motions. Here, we present the cumulative variance of atomic coordinate fluctuations (CVCF) along trajectories as an intuitive PES sensitive metric to assess both the extent of sampling and protein dynamics captured in MD simulations. We first examine the sampling problem in model one- (1D) and two-dimensional (2D) PES to demonstrate that the CVCF when traced as a function of the sampling variable (time in MD simulations) can identify local and global equilibria. Further, even far from global equilibrium, a situation representative of standard MD trajectories of proteins, the CVCF can distinguish different PES and therefore resolve the resultant protein dynamics. We demonstrate the utility of our CVCF analysis by applying it to distinguish the dynamics of structurally homologous proteins from the ubiquitin family (ubiquitin, SUMO1, SUMO2) and ubiquitin protein-protein interactions. Our CVCF analysis reveals that differential side-chain dynamics from the structured part of the protein (the conserved β-grasp fold) present distinct protein PES to distinguish ubiquitin from SUMO isoforms. Upon binding to two functionally distinct protein partners (UBCH5A and UEV), intrinsic ubiquitin dynamics changes to reflect the binding context even though the two proteins have similar binding modes, which lead to negligible (sub-angstrom scale) structural changes.

Machine learning surrogates for molecular dynamics simulations of soft materials

Molecular dynamics (MD) simulations accelerated by high-performance computing (HPC) methods are powerful tools to investigate and extract the microscopic mechanisms characterizing the properties of soft materials such as self-assembled nanoparticles, virus capsids, confined electrolytes, and polymeric fluids. In this paper, we extend the idea developed in our earlier work of integrating machine learning (ML) methods with HPC-accelerated MD simulations of soft materials in order to enhance their predictive power and advance their applications for research and educational activities. Parallelized MD simulations of self-assembling ions in nanoconfinement are employed to demonstrate our approach. We find that an artificial neural network-based regression model successfully learns nearly all the interesting features associated with the output ionic density profiles over a broad range of ionic system parameters. The ML model generates predictions that are in excellent agreement with the results from MD simulations. The inference time associated with the ML model is over a factor of 10,000 smaller than the corresponding parallel MD simulation time. Through this demonstration, we introduce a “machine learning surrogate” for MD simulations of soft-matter systems. We develop and deploy a web application on nanoHUB to realize the advantages associated with the ML surrogate. The results demonstrate that the performance of MD simulations can be further enhanced by using ML, enabling rapid and accurate simulation-driven exploration of the soft material design space.

Revealing the Low-Temperature Fast Relaxation Peak in a Model Metallic Glass

By systematically investigating the relaxation behavior of a model metallic glass based on the extensive molecular dynamics (MD) simulations combined with the dynamic mechanical spectroscopy method, a pronounced ultra-low temperature peak on the loss modulus spectrum was discovered for the first time in MD simulations. It was found that the relation peak occurs at a much lower temperature than the typical temperature for the conventional β relation peak as reported in the literature. According to the atomic displacement analysis, we unravel that the atomic motions, rather than the thermal vibrations or local structural rearrangements, mainly contribute to this relaxation peak. We further identify the atomic level mechanism of this fast relaxation process by characterizing the local geometrical anisotropy. Furthermore, by tracing the dynamic behaviors of these reversible atoms, we demonstrate the intrinsic hierarchy of the local relaxation modes, which are triggered by atomic vibrations and gradually developed to the and irreversible atomic movements. Our findings shed light on a general picture of the relaxation processes in metallic glasses.

Molecular Dynamics Properties without the Full Trajectory: A Denoising Autoencoder Network for Properties of Simple Liquids.

Molecular dynamics (MD) simulation is a popularly used computational tool to compute microscopic and macroscopic properties of a variety of systems including liquids, solids, biological systems, etc. To determine properties of atomic systems to a good level of accuracy with minimal noise or fluctuation, MD simulations are performed over a long time ranging from a few nanoseconds to several tens to hundreds of nanoseconds depending on the system and the properties of interest. In this study, by considering simple liquids, we explore the feasibility of significantly reducing the MD simulation time to compute various properties of monatomic systems such as the structure, pressure, and isothermal compressibility. To do so, extensive MD simulations are performed on 12 000 distinct Lennard-Jones systems at various thermodynamic states. Then, a deep denoising autoencoder network is trained to take the radial distribution function (RDF) from a single snapshot of a Lennard-Jones liquid to compute the mean, temporally averaged RDF. We show that the method is successful in the prediction of RDF and other properties such as the pressure and isothermal compressibility that can be computed based on the RDF not only for Lennard-Jones liquids at various thermodynamic states but also for various simple liquids described by exponential, Yukawa, and inverse-power-law pair potentials.

Molecular Self-Assembly Strategy for Encapsulation of an Amphipathic α-Helical Antimicrobial Peptide into the Different Polymeric and Copolymeric Nanoparticles

Encapsulation of peptide and protein-based drugs in polymeric nanoparticles is one of the fundamental fields in controlled-release drug delivery systems. The molecular mechanisms of absorption of peptides to the polymeric nanoparticles are still unknown, and there is no precise molecular data on the encapsulation process of peptide and protein-based drugs. Herein, the self-assembly of different polymers and block copolymers with combinations of the various molecular weight of blocks and the effects of resultant polymer and copolymer nanomicelles on the stability of magainin2, an α-helical antimicrobial peptide, were investigated by means of all-atom molecular dynamics (MD) simulation. The micelle forming, morphology of micellar aggregations and changes in the first hydration shell of the micelles during micelles formation were explored as well. The results showed that the peptide binds to the polymer and copolymer micelles and never detaches during the MD simulation time. In general, all polymers and copolymers simultaneously encapsulated the peptide during micelles formation and had the ability to maintain the helical structure of the peptide, whereas the first hydration shell of the peptide remained unchanged. Among the micelles, the polyethylene glycol (PEG) micelles completely encapsulated magainin2 and, surprisingly, the NMR structure of the peptide was perfectly kept during the encapsulation process. The MD results also indicated that the aromatic and basic residues of the peptide strongly interact with polymers/copolymers and play important roles in the encapsulation mechanism. This research will provide a good opportunity in the design of polymer surfaces for drug delivery applications such as controlled-release peptide delivery systems.

Molecular-dynamics simulations of solid phase epitaxy in silicon: Effects of system size, simulation time, and ensemble

Solid phase epitaxial (SPE) recrystallization of amorphous Si on a Si(001) substrate was examined by large-scale (6144–129024 Si atoms), long-time (up to 2000 ns) molecular-dynamics (MD) simulations using the empirical Tersoff interatomic potential. We particularly focused on the effects of the MD cell size, simulation time, and ensemble on the SPE growth rate. We found that the simulations under the isothermal–isochoric conditions (NVT ensemble) show a higher crystallization rate than those under the isothermal–isobaric conditions (NPT ensemble). The system size dealt with in the present MD simulation, i.e., >6144 Si atoms, was enough to estimate the SPE growth rate. The Arrhenius plot of the growth rate between 1300 and 1600 K exhibited a single activation energy, ∼2.4 eV, which is in agreement with the experimental value (∼2.7 eV). However, the growth rate at temperatures below 1300 K deviated from the extrapolated ones from 1300 to 1600 K, which is because recrystallization does not reach a steady state: long-time MD simulations are required to estimate the growth rate at low temperature. The structure analysis of amorphous/crystalline interfaces suggested that the braking of atomic bonds parallel to the interface becomes a rate-limiting step of the SPE growth.

Energy Landscape of Negatively Charged BSA Adsorbed on a Negatively Charged Silica Surface.

We study the energy landscape of the negatively charged protein bovine serum albumin adsorbed on a negatively charged silica surface at pH 7. We use fully atomistic molecular dynamics (MD) and steered MD (SMD) to probe the energy of adsorption and the pathway for the surface diffusion of the protein and its associated activation energy. We find an adsorption energy ∼1.2 eV, which implies that adsorption is irreversible even on experimental time scales of hours. In contrast, the activation energy for surface diffusion is ∼0.4 eV so that it is observable on the MD simulation time scale of 100 ns. This analysis paves the way for a more detailed understanding of how a protein layer forms on biomaterial surfaces, even when the protein and surface share the same electrical polarity.

Physical quantity of residue electrostatic energy in flavin mononucleotide binding protein dimer

The electrostatic (ES) energy of each residue was for the first time quantitatively evaluated in a flavin mononucleotide binding protein (FBP). A residue electrostatic energy (RES) was obtained as the sum of the ES energies between atoms in each residue and all other atoms in the FBP dimer using atomic coordinates obtained by a molecular dynamics (MD) simulation. ES is one of the most important energies among the interaction energies in a protein. It is determined from the RES, the residues which mainly contribute to stabilize the structure of each subunit, and the binding energy between two subunits can be estimated. The RES of all residues in subunit A (Sub A) and subunit B (Sub B) were attractive forces, even though the residues contain net negative or positive charges. This reveals that the ES energies of any of the residues can contribute to stabilize the protein structure. The total binding ES energy over all residues among the subunits was distributed between −0.2 to −1.2 eV (mean = −0.67 eV) from the MD simulation time.

Molecular dynamics and MM/GBSA-integrated protocol probing the correlation between biological activities and binding free energies of HIV-1 TAR RNA inhibitors

The interaction of HIV-1 transactivator protein Tat with its cognate transactivation response (TAR) RNA has emerged as a promising target for developing antiviral compounds and treating HIV infection, since it is a crucial step for efficient transcription and replication. In the present study, molecular dynamics (MD) simulations and MM/GBSA calculations have been performed on a series of neamine derivatives in order to estimate appropriate MD simulation time for acceptable correlation between ΔGbind and experimental pIC50 values. Initially, all inhibitors were docked into the active site of HIV-1 TAR RNA. Later to explore various conformations and examine the docking results, MD simulations were carried out. Finally, binding free energies were calculated using MM/GBSA method and were correlated with experimental pIC50 values at different time scales (0–1 to 0–10 ns). From this study, it is clear that in case of neamine derivatives as simulation time increased the correlation between binding free energy and experimental pIC50 values increased correspondingly. Therefore, the binding energies which can be interpreted at longer simulation times can be used to predict the bioactivity of new neamine derivatives. Moreover, in this work, we have identified some plausible critical nucleotide interactions with neamine derivatives that are responsible for potent inhibitory activity. Furthermore, we also provide some insights into a new class of oxadiazole-based back bone cyclic peptides designed by incorporating the structural features of neamine derivatives. On the whole, this approach can provide a valuable guidance for designing new potent inhibitors and modify the existing compounds targeting HIV-1 TAR RNA.

A computational study on the superionic behaviour of ThO2.

This study reports the density functional theory (DFT) and classical molecular dynamics (MD) study of the lattice dynamical, mechanical and anionic transport behaviours of ThO2 in the superionic state. DFT calculations of phonon frequencies were performed at different levels of approximation as a function of isotropic dilation (ε) in the lattice parameter. With the expansion of the lattice parameter, there is a softening of B1u and Eu phonon modes at the X symmetry point of the Brillouin zone. As a result of the nonlinear decrease at the X point, the B1u and Eu phonon modes cross each other at ε = 0.03, which is associated with a sharp increase in the narrow peak of the phonon density of states, signifying a higher occupation and hence a higher coupling of these modes at high temperatures. The mode crossing also indicates anionic conductivity in the 〈001〉 direction leading to occupation of interstitial sites. Moreover, MD and nudged elastic band calculated diffusion barriers indicate that 〈001〉 is the easy direction for anion migration in the normal and superionic states. With a further increase in the lattice parameter, the B1u mode continues to soften and becomes imaginary at a strain (ε) of 0.036 corresponding to a temperature of 3430 K. The calculated temperature variation of single crystal elastic constants shows that the fluorite phase of ThO2 remains elastically stable up to the superionic regime, though the B1u phonon mode is imaginary in that state. This leads to anionic disorder at elevated temperatures. Tracking of anion positions in the superionic state as a function of time in MD simulations suggests a hopping model in which the oxygen ions migrate from one tetrahedral site to another via octahedral interstitial sites.

Molecular Dynamics Study of Helicobacter pylori Urease.

Helicobacter pylori have been implicated in an array of gastrointestinal disorders including, but not limited to, gastric and duodenal ulcers and adenocarcinoma. This bacterium utilizes an enzyme, urease, to produce copious amounts of ammonia through urea hydrolysis in order to survive the harsh acidic conditions of the stomach. Molecular dynamics (MD) studies on the H. pylori urease enzyme have been employed in order to study structural features of this enzyme that may shed light on the hydrolysis mechanism. A total of 400 ns of MD simulation time were collected and analyzed in this study. A wide-open flap state previously observed in MD simulations on Klebsiella aerogenes [Roberts et al. J. Am. Chem. Soc.2012, 134, 9934] urease has been identified in the H. pylori enzyme that has yet to be experimentally observed. Critical distances between residues on the flap, contact points in the closed state, and the separation between the active site Ni2+ ions and the critical histidine α322 residue were used to characterize flap motion. An additional flap in the active site was elaborated upon that we postulate may serve as an exit conduit for hydrolysis products. Finally we discuss the internal hollow cavity and present analysis of the distribution of sodium ions over the course of the simulation.

St-Analyzer: A Web-Based User Interface for Simulation Trajectory Analysis

Molecular dynamics (MD) simulation and its analysis is one of the key tools to provide deeper insights into biological systems at the molecular level, and the results from MD simulation often make critical impacts on defining in-silico models for biological phenomena and validating hypothesis. Recent advances in computing resources and MD simulators significantly accelerate the MD simulation time and thus increase the amount of trajectory data. However, analyzing MD trajectories is still time consuming and often a difficult task even though many laboratories routinely perform MD simulations. ST-analyzer, http://im.bioinformatics.ku.edu/st-analyzer, is a standalone GUI toolset to perform various trajectory analyses with commonly used built-in analysis modules. ST-analyzer has several outstanding features compared to other existing analysis tools: i) handling various formats of trajectory files from CHARMM, NAMD, GROMACS, and Amber, ii) intuitive web-based GUI environment - minimizing administrative load and reducing users' burdens from adapting new software environments, iii) platform independent design scheme - works with any existing OS, iv) easily integrates into job queuing systems - providing options of batch processing either on the cluster or in interactive mode, and v) providing independence between foreground GUI and background modules - makes it easier to add personal modules or to recycle/integrate pre-existing scripts utilizing other analysis tools. The current ST-analyzer contains nine main modules and 18 sub-processors including density profile, lipid deuterium order parameters, lipid surface area, lipid hydrophobic thickness, and more.

Finding inhibitors of mutant superoxide dismutase-1 for amyotrophic lateral sclerosis therapy from traditional chinese medicine.

Superoxide dismutase type 1 (SOD1) mutations cause protein aggregation and decrease protein stability, which are linked to amyotrophic lateral sclerosis (ALS) disease. This research utilizes the world's largest traditional Chinese medicine (TCM) database to search novel inhibitors of mutant SOD1, and molecular dynamics (MD) simulations were used to analyze the stability of protein that interacted with docked ligands. Docking results show that hesperidin and 2,3,5,4′-tetrahydroxystilbene-2-O-β-D-glucoside (THSG) have high affinity to mutant SOD1 and then dopamine. For MD simulation analysis, hesperidin and THSG displayed similar value of RMSD with dopamine, and the migration analysis reveals stable fluctuation at the end of MD simulation time. Interestingly, distance between the protein and ligand has distinct difference, and hesperidin changes the position from initial binding site to the other place. In flexibility of residues analysis, the secondary structure among all complexes does not change, indicating that the structure are not affect ligand binding. The binding poses of hesperidin and THSG are similar to dopamine after molecular simulation. Our result indicated that hesperidin and THSG might be potential lead compound to design inhibitors of mutant SOD1 for ALS therapy.

Hyper-QC: An accelerated finite-temperature quasicontinuum method using hyperdynamics

The quasicontinuum (QC) method is a spatial multiscale method that extends the length scales accessible to fully atomistic simulations (like molecular dynamics (MD)) by several orders of magnitude. While the recent development of the so-called “hot-QC method” enables dynamic simulations at finite temperature, the times accessible to these simulations remain limited to the sub-microsecond time scale due to the small time step required for stability of the numerical integration. To address this limitation, we develop a novel finite-temperature QC method that can treat much longer time scales by coupling the hot-QC method with hyperdynamics—a method for accelerating time in MD simulations. We refer to the new approach as “hyper-QC”. As in the original hyperdynamics method, hyper-QC is targeted at dynamical systems that exhibit a separation of time scales between short atomic vibration periods and long waiting times in metastable states. Acceleration is achieved by modifying the hot-QC potential energy to reduce the energy barriers between metastable states in a manner that ensures that the characteristic dynamics of the system are preserved. First, the high accuracy of hot-QC in reproducing rare event kinetics is demonstrated. Then, the hyper-QC methodology is validated by comparing hyper-QC results with those of hot-QC and full MD for a one-dimensional chain of atoms interacting via a Lennard–Jones potential.

Modeling the Interaction of Dodecylphosphocholine Micelles with the Anticoccidial Peptide PW2 Guided by NMR Data

Antimicrobial peptides are highly dynamic entities that acquire structure upon binding to a membrane interface. To better understand the structure and the mechanism for the molecular recognition of dodecylphosphocholine (DPC) micelles by the anticoccidial peptide PW2, we performed molecular dynamics (MD) simulations guided by NMR experimental data, focusing on strategies to explore the transient nature of micelles, which rearrange on a millisecond to second timescale. We simulated the association of PW2 with a pre-built DPC micelle and with free-DPC molecules that spontaneously forms micelles in the presence of the peptide along the simulation. The simulation with spontaneous micelle formation provided the adequate environment which replicated the experimental data. The unrestrained MD simulations reproduced the NMR structure for the entire 100 ns MD simulation time. Hidden discrete conformational states could be described. Coulomb interactions are important for initial approximation and hydrogen bonds for floating the aromatic region at the interface, being essential for the stabilization of the interaction. Arg9 is strongly attached with phosphate. We observed a helix elongation process stabilized by the intermolecular peptide-micelle association. Full association that mimics the experimental data only happens after complete micelle re-association. Fast micelle dynamics without dissociation of surfactants leads to only superficial binding.

Molecular Dynamics Investigation on a Series of HIV Protease Inhibitors: Assessing the Performance of MM-PBSA and MM-GBSA Approaches

The binding free energies (ΔG(Bind)) obtained from molecular mechanics with Poisson-Boltzmann surface area (MM-PBSA) or molecular mechanics with Generalized Born surface area (MM-GBSA) calculations using molecular dynamics (MD) trajectories are the most popular procedures to measure the strength of interactions between a ligand and its receptor. Several attempts have been made to correlate the ΔG(Bind) and experimental IC(50) values in order to observe the relationship between binding strength of a ligand (with its receptor) and its inhibitory activity. The duration of MD simulations seems very important for getting acceptable correlation. Here, we are presenting a systematic study to estimate the reasonable MD simulation time for acceptable correlation between ΔG(Bind) and experimental IC(50) values. A comparison between MM-PBSA and MM-GBSA approaches is also presented at various time scales. MD simulations (10 ns) for 14 HIV protease inhibitors have been carried out by using the Amber program. MM-PBSA/GBSA based ΔG(Bind) have been calculated and correlated with experimental IC(50) values at different time scales (0-1 to 0-10 ns). This study clearly demonstrates that the MM-PBSA based ΔG(Bind) (ΔG(Bind)-PB) values provide very good correlation with experimental IC(50) values (quantitative and qualitative) when MD simulation is carried out for a longer time; however, MM-GBSA based ΔG(Bind) (ΔG(Bind)-GB) values show acceptable correlation for shorter time of simulation also. The accuracy of ΔG(Bind)-PB increases and ΔG(Bind)-GB remains almost constant with the increasing time of simulation.

THE MOBILITY OF THE EDGE DISLOCATION IN METAL: A MOLECULAR DYNAMICS SIMULATION

In this paper, we use molecular dynamics (MD) simulations and a modified analytic embedded-atom method to investigate the edge dislocation movement without imposed strain at 0 K. The obtained results indicate that the straight lines of the partial dislocations always preserve their original shapes and are parallel to each other during the simulation process. According to the energy of each atom, the positions of both partial dislocation cores are determined. Then the velocities in the period of the relaxation process are investigated in detail. The MD simulations reveal that the MD relaxation time dependence of the edge dislocation mobility is divided into two parts. First, during the initial period ranging from 0 to 6 ps, the relative velocity of the dislocation movement lineally increases with the incremental relaxation time. Second, in the latter period from 6 ps to the end of the simulated process the velocity decreases exponentially as the MD simulation time evolves.