Three‐body expansion of the fragment molecular orbital method combined with density‐functional tight‐binding

Abstract

The three-body fragment molecular orbital (FMO3) method is formulated for density-functional tight-binding (DFTB). The energy, analytic gradient, and Hessian are derived in the gas phase, and the energy and analytic gradient are also derived for polarizable continuum model. The accuracy of FMO3-DFTB is evaluated for five proteins, sodium cation in explicit solvent, and three isomers of polyalanine. It is shown that FMO3-DFTB is considerably more accurate than FMO2-DFTB. Molecular dynamics simulations for sodium cation in water are performed for 100 ps, yielding radial distribution functions and coordination numbers. © 2017 Wiley Periodicals, Inc.

Rapid and accurate assessment of GPCR–ligand interactions Using the fragment molecular orbital‐based density‐functional tight‐binding method

“…We applied FMO code version 5.1 distributed inside ab initio quantum chemistry package GAMESS . We used the third order DFTB3 method with 3ob parameters, and the Møller–Plesset second order perturbation theory (MP2) and, for treating solvent effects, we combined both calculations with the polarizable continuum model (PCM) . MP2 was used with the 6–31G* basis set whereas the UFF dispersion model was used for DFTB3.…”

“…[18] We used the third order DFTB3 method [6] with 3ob parameters, [19,20] and the Møller-Plesset second order perturbation theory (MP2) and, for treating solvent effects, we combined both calculations with the polarizable continuum model (PCM). [4] MP2 was used with the 6-31G* basis set whereas the UFF dispersion model was used for DFTB3. The structures were taken from the previous study.…”

Hierarchical parallelization of divide‐and‐conquer density functional tight‐binding molecular dynamics and metadynamics simulations

“…DFTB‐MD simulations have previously provided encouraging results for various systems including nanomaterials, [ 40,41 ] biomolecules, [ 42,43 ] and condensed phases. [ 43–45 ] The efficiency of DFTB‐MD for large systems has been greatly improved by combination with fragment molecular orbital [ 46–49 ] or divide‐and‐conquer (DC) techniques. [ 50,51 ] Enhanced sampling methods have been recently applied to DFTB‐MD, resulting in novel DFTB‐REMD, [ 52 ] DFTB‐REUS, [ 53,54 ] and DFTB‐MetaD methods.…”

Parallel implementation of efficient charge–charge interaction evaluation scheme in periodic divide‐and‐conquer density‐functional tight‐binding calculations

“…The latter two approaches practically reduce the computational complexity to a nearly linear complexity. Thus, interest in such low‐scaling QM‐MD simulations has become intense …”

“…Thus, interest in such low-scaling QM-MD simulations has become intense. [14][15][16][17][18][19][20][21][22][23][24][25][26] With respect to the above considerations, we recently reported a massively parallel implementation of a divide-andconquer density-functional tight-binding (DC-DFTB) method. [27] The DFTB method [28][29][30][31] can be recognized as the simplified version of the Kohn-Sham DFT.…”