Efficient approach to reactive molecular dynamics with accurate forces

Masahiro Higashi, Donald G Truhlar

Research output: Contribution to journalLetter

15 Citations (Scopus)

Abstract

Density functional theory is a powerful and efficient method for calculating potential energy surfaces for chemical reactions, but its application to complex systems, such as reactions in enzymes, is often prohibitively expensive, even when high-level theory is applied only to a primary subsystem, such as an active site, and when the remaining system is treated by molecular mechanics. Here we show how the combination of multiconfiguration molecular mechanics with charge response kernels can speed up such calculations by three or more orders of magnitude. The resulting method, called electrostatically embedded multiconfiguration molecular mechanics, is illustrated by calculating the free energy of activation profile for the dehalogenation of 1,2-dichloroethane by haloalkane dehalogenase. This shows how hybrid density functionals or other high-level electronic structure methods can now be used efficiently in simulations that require extensive sampling, such as for calculating free energy profiles along a high-barrier reaction coordinate.

Original languageEnglish (US)
Pages (from-to)2925-2929
Number of pages5
JournalJournal of Chemical Theory and Computation
Volume5
Issue number11
DOIs
StatePublished - Nov 1 2009

Fingerprint

Molecular mechanics
Molecular dynamics
molecular dynamics
haloalkane dehalogenase
free energy
Free energy
Dehalogenation
profiles
complex systems
functionals
Potential energy surfaces
enzymes
chemical reactions
potential energy
sampling
activation
density functional theory
electronic structure
Electronic structure
Density functional theory

Cite this

Efficient approach to reactive molecular dynamics with accurate forces. / Higashi, Masahiro; Truhlar, Donald G.

In: Journal of Chemical Theory and Computation, Vol. 5, No. 11, 01.11.2009, p. 2925-2929.

Research output: Contribution to journalLetter

@article{aec41fc8e10d4f7c90a27c9ddeca2501,
title = "Efficient approach to reactive molecular dynamics with accurate forces",
abstract = "Density functional theory is a powerful and efficient method for calculating potential energy surfaces for chemical reactions, but its application to complex systems, such as reactions in enzymes, is often prohibitively expensive, even when high-level theory is applied only to a primary subsystem, such as an active site, and when the remaining system is treated by molecular mechanics. Here we show how the combination of multiconfiguration molecular mechanics with charge response kernels can speed up such calculations by three or more orders of magnitude. The resulting method, called electrostatically embedded multiconfiguration molecular mechanics, is illustrated by calculating the free energy of activation profile for the dehalogenation of 1,2-dichloroethane by haloalkane dehalogenase. This shows how hybrid density functionals or other high-level electronic structure methods can now be used efficiently in simulations that require extensive sampling, such as for calculating free energy profiles along a high-barrier reaction coordinate.",
author = "Masahiro Higashi and Truhlar, {Donald G}",
year = "2009",
month = "11",
day = "1",
doi = "10.1021/ct900301d",
language = "English (US)",
volume = "5",
pages = "2925--2929",
journal = "Journal of Chemical Theory and Computation",
issn = "1549-9618",
publisher = "American Chemical Society",
number = "11",

}

TY - JOUR

T1 - Efficient approach to reactive molecular dynamics with accurate forces

AU - Higashi, Masahiro

AU - Truhlar, Donald G

PY - 2009/11/1

Y1 - 2009/11/1

N2 - Density functional theory is a powerful and efficient method for calculating potential energy surfaces for chemical reactions, but its application to complex systems, such as reactions in enzymes, is often prohibitively expensive, even when high-level theory is applied only to a primary subsystem, such as an active site, and when the remaining system is treated by molecular mechanics. Here we show how the combination of multiconfiguration molecular mechanics with charge response kernels can speed up such calculations by three or more orders of magnitude. The resulting method, called electrostatically embedded multiconfiguration molecular mechanics, is illustrated by calculating the free energy of activation profile for the dehalogenation of 1,2-dichloroethane by haloalkane dehalogenase. This shows how hybrid density functionals or other high-level electronic structure methods can now be used efficiently in simulations that require extensive sampling, such as for calculating free energy profiles along a high-barrier reaction coordinate.

AB - Density functional theory is a powerful and efficient method for calculating potential energy surfaces for chemical reactions, but its application to complex systems, such as reactions in enzymes, is often prohibitively expensive, even when high-level theory is applied only to a primary subsystem, such as an active site, and when the remaining system is treated by molecular mechanics. Here we show how the combination of multiconfiguration molecular mechanics with charge response kernels can speed up such calculations by three or more orders of magnitude. The resulting method, called electrostatically embedded multiconfiguration molecular mechanics, is illustrated by calculating the free energy of activation profile for the dehalogenation of 1,2-dichloroethane by haloalkane dehalogenase. This shows how hybrid density functionals or other high-level electronic structure methods can now be used efficiently in simulations that require extensive sampling, such as for calculating free energy profiles along a high-barrier reaction coordinate.

UR - http://www.scopus.com/inward/record.url?scp=73949147308&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=73949147308&partnerID=8YFLogxK

U2 - 10.1021/ct900301d

DO - 10.1021/ct900301d

M3 - Letter

VL - 5

SP - 2925

EP - 2929

JO - Journal of Chemical Theory and Computation

JF - Journal of Chemical Theory and Computation

SN - 1549-9618

IS - 11

ER -