Assessing the validity of DLPNO‐CCSD(T) in the calculation of activation and reaction energies of ubiquitous enzymatic reactions. Issue 29 (3rd September 2020)

• Record Type: Journal Article
• Authors: Paiva, Pedro; Ramos, Maria J.; Fernandes, Pedro A.
• Is Part Of: Journal of computational chemistry. Volume 41:Issue 29(2020)
• Page Start: 2459
• Page End: 2468
• Publication Date: 2020-09-03
• Subjects: benchmarking -- coupled cluster -- density functional theory -- DLPNO‐CCSD(T) -- quantum mechanics; Chemistry -- Data processing -- Periodicals; 542.85
• DOI: 10.1002/jcc.26401
• Languages: English
• ISSNs: 0192-8651

Abstract

Abstract: The domain‐based local pair natural orbital coupled‐cluster with single, double, and perturbative triples excitation (DLPNO‐CCSD(T)) method was employed to portray the activation and reaction energies of four ubiquitous enzymatic reactions, and its performance was confronted to CCSD(T)/complete basis set (CBS) to assess its accuracy and robustness in this specific field. The DLPNO‐CCSD(T) results were also confronted to those of a set of density functionals (DFs) to understand the benefit of implementing this technique in enzymatic quantum mechanics/molecular mechanics (QM/MM) calculations as a second QM component, which is often treated with DF theory (DFT). On average, the DLPNO‐CCSD(T)/aug‐cc‐pVTZ results were 0.51 kcal·mol −1 apart from the canonic CCSD(T)/CBS, without noticeable biases toward any of the reactions under study. All DFs fell short to the DLPNO‐CCSD(T), both in terms of accuracy and robustness, which suggests that this method is advantageous to characterize enzymatic reactions and that its use in QM/MM calculations, either alone or in conjugation with DFT, in a two‐region QM layer (DLPNO‐CCSD(T):DFT), should enhance the quality and faithfulness of the results. Abstract : The performance of domain‐based local pair natural orbital coupled‐cluster with single, double, and perturbative triples excitation (DLPNO‐CCSD(T)) in portraying the activation and reaction energies of four ubiquitous enzymatic reactions is evaluated and compared to that of CCSD(T)/complete basis set (CBS) and of a set of density functionals (DFs). The DLPNO‐CCSD(T) systematically delivers accurate and robust results that are superior to any of the tested DFs, closely resembling those determined with CCSD(T)/CBS. The authors conclude that DLPNO‐CCSD(T) is adequate to characterize enzymatic reactions and discuss its use within the framework of a quantum mechanics/molecular mechanics calculation.