Estimating the potential energy surface is critical to the study of the thermodynamics and kinetics of environmentally relevant reactions and processes. Umbrella sampling using ab initio molecular dynamics (AIMD) to obtain the potential of mean force (PMF) of a reaction is computationally expensive. One approach to reduce this cost is to combine umbrella sampling and free molecular dynamic simulations and using Discrete Transition-based Reweighting Analysis Method (DTRAM) to estimate the Markov state model and obtain the PMF. The objective of this study is to i) identify the duration of short umbrella and free MD simulations necessary to generate equivalent PMF from 11 ns long umbrella simulations over 21 umbrella windows combined with 1 ns free MD simulations, ii) determine the equilibrium constant of the reaction, and iii) the reaction kinetics. Two separate systems, one with a NaCl molecule and another with a BaSO<sub>4</sub> molecule were investigated in this study. Both systems were solvated with 2000 water molecules. The general Amber Force Field was used to describe the NaCl and BaSO<sub>4</sub> molecular interactions and the SPCE model was used to describe the water. Each of the structures was minimized to remove bad contacts followed by molecular dynamics simulations of heating, constant pressure, and constant volume at 300K. Constant total energy classical thermodynamics was used for the simulations. 21 umbrella windows from 2 to 12 Angstroms spaced at 0.5 Angstroms were used for the umbrella simulations. Each short umbrella simulation was combined with 5 short free MD simulations, both of different time lengths. The results were analyzed with DTRAM using the PyEMMA library. The findings from this study can be used to assess uncertainties from short AIMD or QM/MM/MD simulations. The results showed that there are combinations of short free MD simulations on top of short umbrella simulations that improved the PMF estimation by reducing the standard deviation compared to PMFs with short umbrella simulations only. However, there are large discrepancies between calculated mean first passage times and association constants for short simulations compared to the long 11 ns simulations. This indicates that large uncertainties in estimated quantities using AIMD with short simulations can be expected.

Assessing the potential of Markov state models in reducing the computational expense of estimating PMF, equilibrium constants and kinetics of reactions

Macromolecular Chemistry: The Second Century

[Author Benefits](https://scimeetings.acs.org/?utm_source=pubs_content_marketing&utm_medium=website&utm_campaign=0320_MCF_NPI_Launch_Spring_Homepage&ref=pubs_content_marketing)

[How to use SciMeetings](https://storage.googleapis.com/dl.morressier.com/acs/SciMeetings-ACSSpring2020-HowToGuide_M3020-Final.pdf)

[Frequently Asked Questions](https://storage.googleapis.com/dl.morressier.com/acs/SciMtng_ACSNMSp2020_FAQ_M3120_Final.pdf)

ACS Spring 2020 National Meeting & Expo

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.