The task of accurately conducting free energy simulations is made complicated by two discordant forces: the need for adequate sampling, and the need for accurately describing inter- and intra-molecular interactions, which is prohibitive over long simulations. In this work, the recently developed QM-NBB (Quantum Mechanical Non-Boltzmann Bennett) method has been extended to compute potentials of mean force (PMFs). Initially, a simple conformational change has been modeled – the torsional rotation around butane’s central dihedral – to validate this novel free energy simulation technique. Using this new approach we were able to arrive at QM quality (e.g., HF/6-31G*) results using simulations carried out at the molecular mechanical (MM, CHARMM c36ff) and semi-empirical (SCC-DFTB) levels of theory. This new, “<i>vertical</i>”, QM-NBB method accurately and efficiently calculates perturbations between the underlying MM and QM PMFs. This is accomplished by cleverly assembling the QM end-state ensembles from “intermediate” semi-empirical simulations and then re-evaluating simulation data at the desired Hamiltonian under the influence of a biasing potential appropriate to the differences between the two levels of theory. The data presented from this test case illustrates how well suited this new method is for overcoming costly sampling limitations when highly accurate results are desired.

Overcoming costly sampling simulations using non-Boltzmann Bennet's acceptance ratio method to generate potentials of mean force

249th National Meeting (2015)

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