An efficient and stable hybrid extended Lagrangian/self-consistent field scheme for solving classical mutual induction

Albaugh, Alex; Demerdash, Omar; Head‐Gordon, Teresa

doi:10.1063/1.4933375

Cited by 57 publications

(130 citation statements)

References 31 publications

Supporting

Mentioning

126

Contrasting

Order By: Relevance

“…37 More specifically, the real selfconsistent induced dipoles {µ SCF,I } take the role of {χ SCF,a }, and the auxiliary dipoles {µ I } replace {ζ a } in (10). The nuclear centers R I are propagated in the usual way, i.e.…”

Section: A Adaptation To Classical Polarizable Force-field Methodsmentioning

confidence: 99%

Performance of extended Lagrangian schemes for molecular dynamics simulations with classical polarizable force fields and density functional theory

Vitale

Dziedzic

Albaugh

et al. 2017

The Journal of Chemical Physics

Self Cite

View full text Add to dashboard Cite

1Iterative energy minimization with the aim of achieving self-consistency is a common feature of Born-Oppenheimer molecular dynamics (BOMD) and classical molecular dynamics with polarizable force fields. In the former the electronic degrees of freedom are optimized, while the latter often involves an iterative determination of induced point dipoles. The computational effort of the self-consistency procedure can be reduced by re-using converged solutions from previous time steps. However, this must be done carefully, as not to break time-reversal symmetry, which negatively impacts energy conservation. Self-consistent schemes based on the extended Lagrangian formalism, where the initial guesses for the optimized quantities are treated as auxiliary degrees of freedom, constitute one elegant solution. We report on the performance of two integration schemes with the same underlying extended Lagrangian structure, which we both employ in two radically distinct regimes -in classical molecular dynamics simulations with the AMOEBA polarizable force field, and in BOMD simulations with the Onetep linear-scaling density functional theory (LS-DFT) approach. Both integration schemes are found to offer significant improvements over the standard (unpropagated) molecular dynamics formulation in both the classical and LS-DFT regime.2

show abstract

Section: A Adaptation To Classical Polarizable Force-field Methodsmentioning

confidence: 99%

Performance of extended Lagrangian schemes for molecular dynamics simulations with classical polarizable force fields and density functional theory

Vitale

Dziedzic

Albaugh

et al. 2017

The Journal of Chemical Physics

Self Cite

View full text Add to dashboard Cite

show abstract

“…We note the speedup on GPUs running double precision is ~5 times faster, whereas using mixed precision is roughly 10-20 times faster, than that available with double precision versions of TINKER running OpenMP parallelization on traditional multi-core CPU compute nodes. In addition, speed up doubles when using the iELSCF 72 was used for this test. The ability to run 100ns simulations in a reasonable time on small to medium-sized proteins opens up the possibility of using AMOEBA to perform a wide range of ligand and drug-binding free energy calculations or long biomolecular simulations in which sampling is important, for example for intrinsically disordered proteins.…”

Section: Openmm For Gpu-based Calculations With Amoebamentioning

confidence: 99%

“…We have previously shown 72 that this straight adaptation of Niklasson's method is insufficient for both for decreasing computational expense and maintaining numerical stability due to problems that arise from resonances that make the auxiliary dipoles increasingly poor initial guesses for the SCF solutions of the physical induced dipoles. More specifically, we showed that the auxiliary dipoles evolve on a much faster timescale than their physical counterparts owing to the optimal choice of their characteristic frequency, √2⁄∆t, and their direct coupling in the auxiliary potential (Eq.…”

Section: New Approaches To Solving Mutual Polarizationmentioning

confidence: 99%

“…To address this problem, we introduced "temperature" control on the auxiliary set of dipoles, using both Berendsen 115 rescaling and time-reversible Nosé-Hoover chains 9-10 as thermostating methods, much as is done in Car-Parrinello ab initio MD calculations [116][117] . The set point pseudo-temperature of the auxiliary system can be determined with an equipartition argument applied to the harmonic restraining forces experienced by the auxiliary dipoles 72 .…”

Section: New Approaches To Solving Mutual Polarizationmentioning

confidence: 99%

“…72 Average potential energy and molecular dipole were calculated from NVT simulations at 298.0 K. Diffusion coefficients were averaged over multiple NVE simulation using independent snapshots from 298.0 K NVT simulations as the initial condition. The simulation protocol consisted of a Velocity Verlet integrator with a time step of 1.0 fs, a 9.0 Å vdW cutoff with smoothing, and a 7.0 Å real space cutoff for particle-mesh Ewald (PME) electrostatics.…”

Section: Section 6 Conclusionmentioning

confidence: 99%

See 2 more Smart Citations

Advanced Potential Energy Surfaces for Molecular Simulation

Albaugh¹,

Boateng

Bradshaw

et al. 2016

J. Phys. Chem. B

Self Cite

View full text Add to dashboard Cite

Advanced potential energy surfaces are defined as theoretical models that explicitly include many-body effects that transcend the standard fixed-charge, pairwise-additive paradigm typically used in molecular simulation. However, several factors relating to their software implementation have precluded their widespread use in condensed-phase simulations: the computational cost of the theoretical models, a paucity of approximate models and algorithmic improvements that can ameliorate their cost, underdeveloped interfaces and limited dissemination in computational code bases that are widely used in the computational chemistry community, and software implementations that have not kept pace with modern high-performance computing (HPC) architectures, such as multicore CPUs and modern graphics processing units (GPUs). In this Feature article we review recent progress made in these areas, including well-defined polarization approximations and new multipole electrostatic formulations, novel methods for solving the mutual polarization equations and increasing the MD time step, combining linear scaling electronic structure methods with new QM/MM methods that account for mutual polarization between the two regions, and the greatly improved software deployment of these models and methods onto GPU and CPU hardware platforms. We have now approached an era where multipole-based polarizable force fields can be routinely used to obtain computational results comparable to state-of-the-art density functional theory while reaching sampling statistics that are acceptable when compared to that obtained from simpler fixed partial charge force fields.

show abstract

Advanced models for water simulations

Demerdash

Wang

Head‐Gordon

2017

WIREs Comput Mol Sci

Self Cite

View full text Add to dashboard Cite

Molecular simulations of water using classical, molecular mechanic potential energy functions have enjoyed a 50-year history of development, and much has been learned regarding their parameterization and the essential physics that must be captured in order to reproduce water properties across the phase diagram and across system sizes, from the dimer to the condensed phase. While pairwise-additive force fields using fixed, point charge-based electrostatics have dominated this history owing to computational cost, their limitations in transferability are being recognized, owing particularly to the lack of many-body effects, as well as an inherent difficulty in capturing quantum mechanical effects that become important at short intermolecular separation. This has spurred an impressive development of novel functional forms and parameterization schemes to account for such effects, especially the leading many-body effect of polarization. This review discusses recent efforts in the development of advanced models of water, particularly with regard to important details of their parameterization from quantum mechanical or experimental data, the development of novel functional forms including machine learning-based models, and algorithms that reduce the computational cost of polarization dramatically, permitting them to potentially become competitive with pairwise-additive models as the standby of condensed-phase simulation. These technical developments are appraised based on their ability to impact numerical calculations on water, particularly the condensed phase, and it is hoped that this article provides a clear connection between the essential physics captured by the model and their fitness across a range of environments.

show abstract

An efficient and stable hybrid extended Lagrangian/self-consistent field scheme for solving classical mutual induction

Cited by 57 publications

References 31 publications

Performance of extended Lagrangian schemes for molecular dynamics simulations with classical polarizable force fields and density functional theory

Performance of extended Lagrangian schemes for molecular dynamics simulations with classical polarizable force fields and density functional theory

Advanced Potential Energy Surfaces for Molecular Simulation

Advanced models for water simulations

Contact Info

Product

Resources

About