Benchmarking Electron Densities and Electrostatic Potentials of Proteins from the Three-Partition Frozen Density Embedding Method

Goez A., Neugebauer J.

Research article (journal) | Peer reviewed

Abstract

The fragment-based Three-Partition Frozen Density Embedding (3-FDE) approach [ Jacob, C. R.; Visscher, L. J. Chem. Phys. 2008, 128, 155102 ] is used to generate protein densities and electrostatic potentials, which are critically assessed in comparison to supermolecular Kohn-Sham Density Functional Theory (DFT) results obtained with sophisticated exchange-correlation functionals. The influence of several parameters and user choices is explored with respect to accuracy and reliability. In addition, a recently implemented combination of the 3-FDE scheme with hybrid functionals is applied in production calculations for the first time. We demonstrate that the 3-FDE method not only closely reproduces results from corresponding supermolecular calculations for routine situations (peptides/proteins in solution) but can even surpass conventional Kohn-Sham DFT in accuracy for difficult cases, such as zwitterionic structures in vacuo. This is due to the fact that the fragmentation inherently limits the overdelocalization caused by the self-interaction error in common DFT approximations. The method is thus not only able to reduce the computational effort for the description of large biological entities but also can strongly reduce the artifacts brought about by the SIE.

Details about the publication

JournalJournal of Chemical Theory and Computation
Volume12
Issue10
Page range4843-4855
StatusPublished
Release year2016
Language in which the publication is writtenEnglish
DOI10.1021/acs.jctc.6b00590
Link to the full texthttps://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=84991042672&origin=inward

Authors from the University of Münster

Goez, Albrecht
Professur für Theoretische Organische Chemie (Prof. Neugebauer)
Neugebauer, Johannes
Professur für Theoretische Organische Chemie (Prof. Neugebauer)
Center for Multiscale Theory and Computation