RNA molecules are challenging to model with molecular mechanical force fields, in part due to the highly charged residues and multitude of �non-canonical� structural motifs.  Recent improvements to RNA force fields have focused on refinements to the torsion parameters by fitting to highly accurate QM data.  While these updates provide overall improvement in the quality of RNA modeling, we have observed some negative side effects, most recently in the sugar pucker representation.   We anticipate that future refinement efforts will require simultaneous fitting of multiple parameters at once.  Toward this goal, we have generated a large set of QM data for mono- and dinucleotides covering a wide range of likely conformations.  New torsional parameters were developed by fitting to a combined QM/MM dataset using genetic search and minimization algorithms.  We will discuss the performance of the new parameters for a variety of RNA simulations.

Simultaneous fitting of RNA torsion parameters to a large QM/MM dataset

243rd National Meeting (2012)

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