molecular dynamics

molecular dynamics
Knordlun · Public domain · source
Name	Molecular dynamics
Developer	John Adams, Alder and Wainwright
Initial release	1957
Genre	Computational simulation

Contents

molecular dynamics

Molecular dynamics is a computational technique for simulating the time evolution of interacting particles using numerical integration of classical equations of motion. Developed alongside efforts in Los Alamos National Laboratory, Princeton University, and IBM, the approach connects atomistic models to observable properties measured in experiments at facilities like Lawrence Livermore National Laboratory and Brookhaven National Laboratory. It underpins studies that inform projects at National Institutes of Health, European Molecular Biology Laboratory, and industrial labs such as Pfizer and DuPont.

Introduction

Molecular dynamics traces roots to early work by researchers at Argonne National Laboratory and collaborations involving scientists associated with Ames Laboratory, University of California, Berkeley, and Stanford University. Applications range from interpretations of data from the Large Hadron Collider to molecular design efforts in partnerships with Merck and Roche. Key historical milestones occurred in conferences hosted by American Physical Society and Royal Society meetings where methods were critiqued alongside developments in supercomputing at Lawrence Berkeley National Laboratory and algorithmic advances showcased at ACM SIGGRAPH.

Algorithmic choices in simulations often reference implementations in software developed at institutions such as University of Illinois Urbana-Champaign, Max Planck Society, and Los Alamos National Laboratory. Integrators like those introduced in work funded by National Science Foundation and presented at SIAM meetings are paired with neighbor-list schemes inspired by projects at Oak Ridge National Laboratory. Parallelization strategies leverage middleware from Cray Research and architectures designed by Intel and NVIDIA. Enhanced-sampling algorithms derived from contributions at Columbia University and University of Cambridge include umbrella sampling, metadynamics, and replica exchange methods promoted at workshops organized by Gordon Research Conferences.

Force-field development stems from collaborations among groups at Harvard University, Yale University, and Massachusetts Institute of Technology producing parameter sets mirrored in packages maintained by Schrödinger (company), OpenMM teams, and communities around GROMACS at Uppsala University. Classical potentials reference foundational datasets curated by researchers affiliated with National Institute of Standards and Technology and comparisons reported in journals published by Nature Publishing Group and American Chemical Society. Polarizable models and coarse-grained potentials have been advanced through consortia involving European Molecular Biology Laboratory and industrial partners like GlaxoSmithKline.

Preparation workflows are taught in courses at Massachusetts Institute of Technology, University of Oxford, and University of Tokyo and implemented in tools from groups at Lawrence Livermore National Laboratory and Sandia National Laboratories. Choices of boundary conditions, thermostats, barostats, and time steps reflect standards discussed at meetings of the Royal Society of Chemistry and validation benchmarks established by National Institutes of Health collaborations. Systems are often equilibrated using procedures validated in studies conducted at Cold Spring Harbor Laboratory and data repositories maintained by European Bioinformatics Institute.

Analyses translate trajectories into observables compared against experiments at California Institute of Technology, Max Planck Institute for Biophysical Chemistry, and Stanford Synchrotron Radiation Lightsource. Applications include ligand docking studies used by Pfizer and Novartis, materials modeling relevant to Boeing and General Electric, and investigations of protein folding that cite experiments from Rosalind Franklin Institute and Scripps Research. Multiscale workflows integrate inputs from projects at Sandia National Laboratories and Los Alamos National Laboratory to inform protocols adopted in collaborations with World Health Organization and policy discussions at United Nations forums.

Limitations and error sources are documented in reviews from journals published by Elsevier and articulated at symposia hosted by American Chemical Society and Biophysical Society. Common issues include finite-size effects noted in benchmark studies from Oak Ridge National Laboratory, force-field transferability concerns raised by researchers at Imperial College London, and sampling insufficiencies discussed in workshops organized at European Academy of Sciences. Validation against experimental data from National Synchrotron Light Source and replication efforts at Cold Spring Harbor Laboratory are essential to quantify uncertainties and model limitations.