How Reproducible are QM/MM Simulations? Lessons from Computational Studies of the Covalent Inhibition of the SARS-CoV-2 Main Protease by Carmofur

Abstract

This work explores what level of transparency in reporting the details is required for practical reproducibility of quantum mechanics/molecular mechanics (QM/MM) simulations. Using the reaction of an essential SARS-CoV-2 enzyme (the main protease) with a covalent inhibitor (carmofur) as a test case of chemical reactions in biomolecules, we carried out QM/MM calculations to determine the structures and energies of the reactants, the product, and the transition state/intermediate using analogous QM/MM models implemented in two software packages, NWChem and QChem. Our main benchmarking goal was to reproduce the key energetics computed with the two packages. Our results indicate that quantitative agreement (within the numerical thresholds used in calculations) is difficult to achieve. We show that rather minor details of QM/MM simulations must be reported in order to ensure the reproducibility of the results and offer suggestions towards developing practical guidelines for the reporting results of biosimulations.

Computational Modeling of the Interaction of Molecular Oxygen with the Flavin-dependent Enzyme RutA

“…As shown in the very recent study [11], the treatment of the QM-MM boundary as well as of the embedding scheme is still an open question despite years of successful application of QM/MM codes in modern software packages. It seems that the only way to report results of QM/MM application that can be reproduced in subsequent simulations is to provide as many technical details as possible.…”