Simple and effective application of the Wang-Landau method for multicanonical molecular dynamics simulation

We propose a novel application of the Wang-Landau method (WLM) for multicanonical molecular dynamics (McMD) simulations. Originally, WLM was developed for Monte Carlo (MC) simulations. Fundamentally, WLM remarkably reduces simulation efforts because it estimates the optimal multicanonical energy function automatically. When WLM is applied to McMD, not only the multicanonical energy but also energy gradient must be estimated adequately. However, because of the rugged multicanonical energy function at the early simulation stage, applications of WLM for MD simulations are difficult and require a smoothing procedure: simulation efforts such as cubic-spline extrapolation and gathering multiple preruns are utilized for smoothing. We propose a simple and effective smoothing method that requires only one additional equation and two time-dependent parameters. As a result, our method produced the correct multicanonical energy function and succeeded in the flat sampling of a small biomolecule with reduced simulation effort.     

Measure of accuracy for multicanonical molecular-dynamics simulation

A measure of the flatness of the energy probability distribution for multicanonical molecular-dynamics (MMD) simulation is presented. It (the flatness measure) can be introduced by a slight change in the renewing scheme of the MMD potential energy. Our proposed measure is applied to liquid Ar with a Lennard-Jones potential system in order to investigate the influence of flatness on the simulation results such as internal energy and specific heat at constant volume. We find that the accuracy of MMD simulation is influenced not only by the flatness of the energy probability distribution but also by the width of the energy region that is accessible during the MMD simulation.     

Optimization of partial multicanonical molecular dynamics simulations applied to an alanine dipeptide in explicit water solvent

The partial multicanonical algorithm for molecular dynamics and Monte Carlo simulations samples a wide range of an important part of the potential energy. Although it is a strong technique for structure prediction of biomolecules, the choice of the partial potential energy has not been optimized. In order to find the best choice, partial multicanonical molecular dynamics simulations of an alanine dipeptide in explicit water solvent were performed with 15 trial choices for the partial potential energy. The best choice was found to be the sum of the electrostatic, Lennard-Jones, and torsion-angle potential energies between solute atoms. In this case, the partial multicanonical simulation sampled all of the local-minimum free-energy states of the P(II), C(5), α(R), α(P), α(L), and C states and visited these states most frequently. Furthermore, backbone dihedral angles ? and ψ rotated very well. It is also found that the most important term among these three terms is the electrostatic potential energy and that the Lennard-Jones term also helps the simulation to overcome the steric restrictions. On the other hand, multicanonical simulation sampled all of the six states, but visited these states fewer times. Conventional canonical simulation sampled only four of the six states: The P(II), C(5), α(R), and α(P) states.     

A model electronic Hamiltonian to study low-lying electronic states of [Fe(bpy)3]2+ in aqueous solution

A simple model electronic Hamiltonian to describe the potential energy surfaces of several low-lying d-d states of the [Fe(bpy)(3)](2+) complex is developed for use in molecular dynamics (MD) simulation studies. On the basis of a method proposed previously for first-row transition metal ions in aqueous solution, the model Hamiltonian is constructed using density functional theory calculations for the lowest singlet and quintet states. MD simulations are then carried out for the two spin states in aqueous solution in order to examine the performance of the model Hamiltonian. The simulation results indicate that the present model electronic Hamiltonian reasonably describes the potential energy surfaces of the two spin states of the aqueous [Fe(bpy)(3)](2+) system, while retaining sufficient simplicity for application in simulation studies on excited state dynamics.     

Prediction of peptide conformation by multicanonical algorithm: New approach to the multiple-minima problem

We apply a recently developed method, the multicanonical algorithm, to the problem of tertiary structure prediction of peptides and proteins. As a simple example to test the effectiveness of the algorithm, metenkephalin is studied and the ergodicity problem, or multiple-minima problem, is shown to be overcome by this algorithm. The lowest-energy conformation obtained agrees with that determined by other efficient methods such as Monte Carlo simulated annealing. The superiority of the present method to simulated annealing lies in the fact that the relationship to the canonical ensemble remains exactly controlled. Once the multicanonical parameters are determined, only one simulation run is necessary to obtain the lowest-energy conformation and further the results of this one run can be used to calculate various thermodynamic quantities at any temperature. The latter point is demonstrated by the calculation of the average potential energy and specific heat as functions of temperature. © John Wiley & Sons, Inc.     

An integrate-over-temperature approach for enhanced sampling

A simple method is introduced to achieve efficient random walking in the energy space in molecular dynamics simulations which thus enhances the sampling over a large energy range. The approach is closely related to multicanonical and replica exchange simulation methods in that it allows configurations of the system to be sampled in a wide energy range by making use of Boltzmann distribution functions at multiple temperatures. A biased potential is quickly generated using this method and is then used in accelerated molecular dynamics simulations.     

Communication: Quantum polarized fluctuating charge model: a practical method to include ligand polarizability in biomolecular simulations

We present a simple and practical method to include ligand electronic polarization in molecular dynamics (MD) simulation of biomolecular systems. The method involves periodically spawning quantum mechanical (QM) electrostatic potential (ESP) calculations on an extra set of computer processors using molecular coordinate snapshots from a running parallel MD simulation. The QM ESPs are evaluated for the small-molecule ligand in the presence of the electric field induced by the protein, solvent, and ion charges within the MD snapshot. Partial charges on ligand atom centers are fit through the multi-conformer restrained electrostatic potential (RESP) fit method on several successive ESPs. The RESP method was selected since it produces charges consistent with the AMBER/GAFF force-field used in the simulations. The updated charges are introduced back into the running simulation when the next snapshot is saved. The result is a simulation whose ligand partial charges continuously respond in real-time to the short-term mean electrostatic field of the evolving environment without incurring additional wall-clock time. We show that (1) by incorporating the cost of polarization back into the potential energy of the MD simulation, the algorithm conserves energy when run in the microcanonical ensemble and (2) the mean solvation free energies for 15 neutral amino acid side chains calculated with the quantum polarized fluctuating charge method and thermodynamic integration agree better with experiment relative to the Amber fixed charge force-field.     

Multicanonical Monte Carlo calculation of the free‐energy map of the base–amino acid interaction

The specific interactions between base pairs and amino acids were studied by the multicanonical Monte Carlo method. We sampled numerous interaction configurations and side‐chain conformations of the amino acid by the multicanonical algorithm, and calculated the free energies of the interactions between an amino acid at given C_α positions and a fixed base pair. The contour maps of free energy derived from this calculation represent the preferred C_α position of the amino acid around the base, and these maps of various combinations of bases and amino acids can be used to quantify the specificity of intrinsic base–amino acid interactions. Similarly, enthalpy and entropy maps will provide further details of the specific interactions. We have also calculated the free‐energy map of the orientations of the C_α C_β bond vector, which indicates the preferential orientation of the amino acid against the base. We compared the results obtained by the multicanonical method with those of the exhaustive sampling and canonical Monte Carlo methods. The free‐energy map of the base–amino acid interaction obtained by the multicanonical simulation method was nearly identical to the accurate result derived from the exhaustive sampling method. This indicates that a single multicanonical Monte Carlo simulation can produce an accurate free‐energy map. Multicanonical Monte Carlo sampling produced free‐energy maps that were more accurate than those produced by canonical Monte Carlo sampling. Thus, the multicanonical Monte Carlo method can serve as a powerful tool for estimating the free‐energy landscape of base–amino acid interactions and for elucidating the mechanism by which amino acids of proteins recognize particular DNA base pairs. © 2000 John Wiley & Sons, Inc. J Comput Chem 21: 954–962, 2000     

Repeated‐annealing sampling combined with multicanonical algorithm for conformational sampling of bio‐molecules

A novel conformational sampling method (repeated‐annealing sampling method) is proposed to execute an efficient conformational sampling at a reasonable computational cost. In the method, a molecular dynamics simulation is done with repeating an elemental process. An elemental process consists of four subprocesses: high‐temperature run, annealing, room‐temperature run, and fast heating. The sampling is done automatically according to a temperature‐control schedule. The room‐temperature run is treated with the multicanonical algorithm, and the other subprocesses are done with the conventional molecular dynamics algorithm. The method, differing from the generalized ensemble methods recently developed, is not warrantable to give the canonical ensemble because of the nonphysical process in the annealing. However, we observed that the slower the annealing and the longer the high‐temperature run, the closer the sampled conformations to those of the canonical ensemble. A test was performed with tri‐N‐acetyl‐D ‐glucosamine in vacuo, and the results were compared with those from the conventional multicanonical simulation. Not only the reweighted canonical distribution function but also the energy landscape were in good agreement with those from the conventional multicanonical simulation. The potential of mean force also showed a fairly good agreement with that from the conventional multicanonical simulation in the room‐temperature region. © 2001 John Wiley & Sons, Inc. J Comput Chem 22: 1098–1106, 2001     

New approach to the first-order phase transition of Lennard-Jones fluids

The multicanonical Monte Carlo method is applied to a bulk Lennard-Jones fluid system to investigate the liquid-solid phase transition. We take the example of a system of 108 argon particles. The multicanonical weight factor we determined turned out to be reliable for the energy range between -7.0 and -4.0 kJ/mol, which corresponds to the temperature range between 60 and 250 K. The expectation values of the thermodynamic quantities obtained from the multicanonical production run by the reweighting techniques exhibit the characteristics of first-order phase transitions between liquid and solid states around 150 K. The present study reveals that the multicanonical algorithm is particularly suitable for analyzing the transition state of the first-order phase transition in detail.     

Enhanced sampling via strong coupling to a heat bath: relationship between Tsallis and multicanonical algorithms

We show that Tsallis and multicanonical statistical mechanics are equivalent under specific conditions and that they describe a system strongly coupled to a heat bath. The concept of the strong coupling to a heat bath, in which energy fluctuation is larger than that in the canonical ensemble [J. Chem. Phys. 119, 7075 (2003)], plays a key role in relating Tsallis formalism to multicanonical formalism. The equivalence between these formalisms allows us to obtain an appropriate q parameter in the Tsallis algorithm to enhance the sampling in the phase space in a manner similar to the multicanonical algorithm. An enhanced sampling in the configurational space by use of the strong coupling formalism is demonstrated in a Lennard-Jones fluid.     

Validation of intermolecular pair potential model of SiH4: molecular-dynamics simulation for saturated liquid density and thermal transport properties

We demonstrate a validation of the intermolecular pair potential model of SiH(4), which is constructed from ab initio molecular-orbital calculations and expressed as the sum of the exponential and the London dispersion terms. The saturated liquid densities of SiH(4) are calculated for temperatures from 100 to 225 K by molecular-dynamics (MD) simulation. The average deviation between the experiment and the MD simulation using the present potential model is 3.9%, while the deviations exceed 10% for other well-known potential models such as the five-center Lennard-Jones (LJ) model. Subsequently, the shear viscosity, the thermal conductivity, and the self-diffusion coefficient of liquid SiH(4) are calculated by an equilibrium MD simulation with the Green-Kubo formula from 100 to 225 K. The average deviations from experiment are 11.8% and 13.7% for the shear viscosity and the thermal conductivity, respectively. Comparing the present model with an empirical one-center LJ model, it turns out that the rotational energy transfer through the intermolecular potential energy, which comes from the anisotropic potential energy, plays an important role in the thermal conductivity of liquid SiH(4). These results indicate that the present intermolecular potential model has an ability to give realistic pictures for liquid SiH(4) through molecular simulations.     

Conformational transition free energy profiles of an adsorbed, lattice model protein by multicanonical Monte Carlo simulation

Proteins often undergo changes in internal conformation upon interacting with a surface. We investigate the thermodynamics of surface induced conformational change in a lattice model protein using a multicanonical Monte Carlo method. The protein is a linear heteropolymer of 27 segments (of types A and B) confined to a cubic lattice. The segmental order and nearest neighbor contact energies are chosen to yield, in the absence of an adsorbing surface, a unique 3x3x3 folded structure. The surface is a plane of sites interacting either equally with A and B segments (equal affinity surface) or more strongly with the A segments (A affinity surface). We use a multicanonical Monte Carlo algorithm, with configuration bias and jump walking moves, featuring an iteratively updated sampling function that converges to the reciprocal of the density of states 1/Omega(E), E being the potential energy. We find inflection points in the configurational entropy, S(E)=k ln Omega(E), for all but a strongly adsorbing equal affinity surface, indicating the presence of free energy barriers to transition. When protein-surface interactions are weak, the free energy profiles F(E)=E-TS(E) qualitatively resemble those of a protein in the absence of a surface: a free energy barrier separates a folded, lowest energy state from globular, higher energy states. The surface acts in this case to stabilize the globular states relative to the folded state. When the protein surface interactions are stronger, the situation differs markedly: the folded state no longer occurs at the lowest energy and free energy barriers may be absent altogether.     

Molecular dynamics simulation in the grand canonical ensemble

An extended system Hamiltonian is proposed to perform molecular dynamics (MD) simulation in the grand canonical ensemble. The Hamiltonian is similar to the one proposed by Lynch and Pettitt (Lynch and Pettitt, J Chem Phys 1997, 107, 8594), which consists of the kinetic and potential energies for real and fractional particles as well as the kinetic and potential energy terms for material and heat reservoirs interacting with the system. We perform a nonlinear scaling of the potential energy parameters of the fractional particle, as well as its mass to vary the number of particles dynamically. On the basis of the equations of motion derived from this Hamiltonian, an algorithm has been proposed for MD simulation at constant chemical potential. The algorithm has been tested for the ideal gas, for the Lennard-Jones fluid over a wide range of temperatures and densities, and for water. The results for the low-density Lennard-Jones fluid are compared with the predictions from a truncated virial equation of state. In the case of the dense Lennard-Jones fluid and water our predicted results are compared with the results reported using other available methods for the calculation of the chemical potential. The method is also applied to the case of vapor-liquid coexistence point predictions.     

Calculations of the free energy of interaction of the c-Fos-c-Jun coiled coil: effects of the solvation model and the inclusion of polarization effects

The leucine zipper region of activator protein-1 (AP-1) comprises the c-Jun and c-Fos proteins and constitutes a well-known coiled coil protein-protein interaction motif. We have used molecular dynamics (MD) simulations in conjunction with the molecular mechanics/Poisson-Boltzmann generalized-Born surface area [MM/PB(GB)SA] methods to predict the free energy of interaction of these proteins. In particular, the influence of the choice of solvation model, protein force field, and water potential on the stability and dynamic properties of the c-Fos-c-Jun complex were investigated. Use of the AMBER polarizable force field ff02 in combination with the polarizable POL3 water potential was found to result in increased stability of the c-Fos-c-Jun complex. MM/PB(GB)SA calculations revealed that MD simulations using the POL3 water potential give the lowest predicted free energies of interaction compared to other nonpolarizable water potentials. In addition, the calculated absolute free energy of binding was predicted to be closest to the experimental value using the MM/GBSA method with independent MD simulation trajectories using the POL3 water potential and the polarizable ff02 force field, while all other binding affinities were overestimated.     

An efficient method for computing excess free energy of liquid

We present a new theoretical method for efficient calculation of free energy of liquid. This interaction entropy method allows one to compute entropy and free energy of liquid from standard single step MD (molecular dynamics) simulation directly in liquid state without the need to perform MD simulations at many intermediate states as required in thermodynamic integration or free energy perturbation methods. In this new approach, one only needs to evaluate the interaction energy of a single (fixed) liquid molecule with the rest of liquid molecules as a function of time from a standard MD simulation of liquid and the fluctuation of distribution of this interaction energy is then used to calculate the interaction entropy of the liquid. Explicit theoretical derivation of this interaction entropy approach is provided and numerical calculations for the benchmark liquid water system were carried out using three different water models. Numerical analysis of the result was performed and comparison of the computational result with experimental data and other theoretical results were provided. Excellent agreement of calculated free energies with the experimental data using TIP4P model is obtained for liquid water.     

Molecular dynamics study of CaCl2 in methanol

The effective potential to describe interactions between the calcium ion and methanol molecule was derived from a potential energy surface. The energy surface was constructed from more than 10,000 points obtained from ab initio calculations, performed at the MP2/6-31G(d,p) level using Gaussian98. The derived effective potential and the Palinkas-Hawlicka-Heinzinger methanol model were employed in the molecular dynamics (MD) simulation of 0.25 M CaCl(2) solution in methanol at 298 K. Average distances between ions and the methanol sites--oxygen and carbon atoms--obtained from MD simulation agreed excellently with the X-ray data. The coordination number of the cation, greater as compared with the experimental results, is in accord with its concentration dependence.     

Comparison between self-guided Langevin dynamics and molecular dynamics simulations for structure refinement of protein loop conformations

This article presents a comparative analysis of two replica‐exchange simulation methods for the structure refinement of protein loop conformations, starting from low‐resolution predictions. The methods are self‐guided Langevin dynamics (SGLD) and molecular dynamics (MD) with a Nosé–Hoover thermostat. We investigated a small dataset of 8‐ and 12‐residue loops, with the shorter loops placed initially from a coarse‐grained lattice model and the longer loops from an enumeration assembly method (the Loopy program). The CHARMM22 + CMAP force field with a generalized Born implicit solvent model (molecular‐surface parameterized GBSW2) was used to explore conformational space. We also assessed two empirical scoring methods to detect nativelike conformations from decoys: the all‐atom distance‐scaled ideal‐gas reference state (DFIRE‐AA) statistical potential and the Rosetta energy function. Among the eight‐residue loop targets, SGLD out performed MD in all cases, with a median of 0.48 Å reduction in global root‐mean‐square deviation (RMSD) of the loop backbone coordinates from the native structure. Among the more challenging 12‐residue loop targets, SGLD improved the prediction accuracy over MD by a median of 1.31 Å, representing a substantial improvement. The overall median RMSD for SGLD simulations of 12‐residue loops was 0.91 Å, yielding refinement of a median 2.70 Å from initial loop placement. Results from DFIRE‐AA and the Rosetta model applied to rescoring conformations failed to improve the overall detection calculated from the CHARMM force field. We illustrate the advantage of SGLD over the MD simulation model by presenting potential‐energy landscapes for several loop predictions. Our results demonstrate that SGLD significantly outperforms traditional MD in the generation and populating of nativelike loop conformations and that the CHARMM force field performs comparably to other empirical force fields in identifying these conformations from the resulting ensembles. Published 2011 Wiley Periodicals, Inc. J Comput Chem, 2011     

Convergence of Free Energy Profile of Coumarin in Lipid Bilayer

Atomistic molecular dynamics (MD) simulations of druglike molecules embedded in lipid bilayers are of considerable interest as models for drug penetration and positioning in biological membranes. Here we analyze partitioning of coumarin in dioleoylphosphatidylcholine (DOPC) bilayer, based on both multiple, unbiased 3 μs MD simulations (total length) and free energy profiles along the bilayer normal calculated by biased MD simulations (～7 μs in total). The convergences in time of free energy profiles calculated by both umbrella sampling and z-constraint techniques are thoroughly analyzed. Two sets of starting structures are also considered, one from unbiased MD simulation and the other from "pulling" coumarin along the bilayer normal. The structures obtained by pulling simulation contain water defects on the lipid bilayer surface, while those acquired from unbiased simulation have no membrane defects. The free energy profiles converge more rapidly when starting frames from unbiased simulations are used. In addition, z-constraint simulation leads to more rapid convergence than umbrella sampling, due to quicker relaxation of membrane defects. Furthermore, we show that the choice of RESP, PRODRG, or Mulliken charges considerably affects the resulting free energy profile of our model drug along the bilayer normal. We recommend using z-constraint biased MD simulations based on starting geometries acquired from unbiased MD simulations for efficient calculation of convergent free energy profiles of druglike molecules along bilayer normals. The calculation of free energy profile should start with an unbiased simulation, though the polar molecules might need a slow pulling afterward. Results obtained with the recommended simulation protocol agree well with available experimental data for two coumarin derivatives.