Molecular dynamics (MD) simulation is a powerful tool used to probe the physical properties of small and large molecular systems. However, the validity of these simulations is ultimately dependent on the use of an accurate force field. The current state-of-the-art force fields (FFs) that are employed in MD simulations (e.g., AMBER, OPLS) are empirical ones with parameters fitted to reproduce a combination of available experimental data. In many cases (e.g., drug discovery), molecules are encountered for which FF parameters are missing. For such molecules, first-principles FFs may be derived by defining an interatomic potential and fitting parameters to quantum mechanical (QM) data. Since FF parameterization involves performing QM calculations, this task scales prohibitively with system size. Here we present results from an automatic, fragment-based application of symmetry-adapted perturbation theory (SAPT) to derive molecular FFs extensible to large systems such as proteins. Three main systems were considered in the present work: the 5,5′-dinitro-2H,2H′-3,3′-bi-1,2,4-triazole (DNBT), ethylene glycol, and ethane dimers. Optimized monomers were split into symmetric fragments and broken valences capped with a hydrogen atom. All SAPT calculations were performed with SAPT2016 using the SAPT(DFT) method with the PBE0 hybrid functional and aug-cc-pVDZ basis set. The autoPES package was used to fit all potentials to SAPT(DFT) data. In all cases, fragment reconstructed FFs (rPESs) had an RMSE from SAPT(DFT) interaction energies of approximately 2x greater than that of the corresponding direct fit PES (dPES) in the attractive regime, while OPLS had such RMSEs ranging from 3x to 6x. RMSEs in the repulsive regime were found to be at least one order of magnitude higher for OPLS than for dPES and rPES. Similar results have also been obtained for the glycine dimer and alanine dipeptide dimer, indicating that the rPESs provide a much more accurate description of the noncovalent interactions of peptides than OPLS as well.

Force fields automatically constructed from ab Initio derived interaction potentials for fragments

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.