Hybrid classical quantum force field for modeling very large molecules

Abstract

mA coherent computational scheme on a very large molecule in which the subsystem that undergoes the most important electronic changes is treated by a semiempirical quantum chemical method, though the rest of the molecule is described by a classical force field, has been proposed recently. The continuity between the two subsystems is obtained by a strictly localized bond orbital, which is assumed to have transferable properties determined on model molecules. The computation of the forces acting on the atoms is now operating, giving rise to a hybrid classical quantum force field (CQFF) which allows full energy minimization and modeling chemical changes in large biomolecules. As an illustrative example, we study the short hydrogen bonds and the proton-exchange process in the histidine-aspartic acid system of the catalytic triad of human neutrophil elastase. The CQFF approach reproduces the crystallographic data quite well, in opposition to a classical force field. The method also offers the possibility of switching off the electrostatic interaction between the quantum and the classical subsystems, allowing us to analyze the various components of the perturbation exerted by the macromolecule in the reactive part. Molecular dynamics confirm a fast proton exchange between the three possible energy wells. The method appears to be quite powerful and applicable to other cases of chemical interest such as surface reactivity of nonmetallic solids.0 1996 John cations of these systems are usually out of reach of these methods in which the basic concept is that of well-established chemical bonds. A chemical reaction path can be computed provided that the system is considered as an assembly of nuclei and electrons and treated quantum mechanically. This constraint limits reactivity studies to small sys-

Electrostatic properties in the catalytic site of papain: A possible regulatory mechanism for the reactivity of the ion pair

“…It can also be used as a first approach in the investigation of which aspects should be retained and investigated in theoretical studies using more sophisticated and more computationally expensive methodologies such as quantum/classical hybrid methods. [85][86][87] …”

QM/MM: what have we learned, where are we, and where do we go from here?

“…The second class of QM/MM methods consists of methods that use localized orbitals at the boundary between the PS and SS. An example is the so-called local self-consistent field (LSCF) algorithm [35,38,43,51,112], where the bonds connecting the PS and SS are represented by a set of strictly localized bond orbitals (SLBOs) that are determined by calculations on small model compounds and assumed to be transferable. The SLBOs are excluded from the self-consistent field (SCF) optimization of the large molecule to prevent their admixture with other QM basis functions.…”

Variational optimization of effective atom centered potentials for molecular properties

“…33 Alternatively, the frontier orbitals of the link atom which point into the direction of the MM region can be considered to be frozen. 34,35 In the present study, we propose to optimize systematically the capping ECPs at the boundaries using density functional perturbation theory. Specifically, using perturbed electron densities we will show how to iteratively minimize differences in electron density between the QM region and an all-atom QM reference calculation.…”