This subproject is one of many research subprojects utilizing theresources provided by a Center grant funded by NIH/NCRR. The subproject andinvestigator (PI) may have received primary funding from another NIH source,and thus could be represented in other CRISP entries. The institution listed isfor the Center, which is not necessarily the institution for the investigator.Biological simulation with empirical force-fields is widely used to rationalize experimental data and generate hypotheses regarding molecular function. Despite its wide-spread use, the accuracy of empirical force-fields for ab-initio folding is still in question. Faster molecular dynamics codes combined with sampling algorithms such as simulated tempering has made it possible to observe unbiased folding of small peptides and proteins and a precise determination of native-state stability. For example, recent parallel tempering simulations by the PI has determined the native structure and stability of the 20-residue trpcage protein. It is necessary to determine if current force-fields are capable of generating physically accurate native-structures for other proteins. In particular, the most popular empirical force-fields have been shown insufficient to simultaneously stabilize the native structures of helical and beta-sheet peptides (C-peptide and the N-terminal hairpin of GB1). Furthermore, wide variations are found between different force-field parameterizations and solvent treatments. The PI is completing a simulation of a small alpha-helical protein - the 37-residue chicken villin headpiece - using simulated tempering and the OPLS/AA force-field in explicit (TIP4P) solvent. Although the OPLSA/AA force-field is known to have a slight propensity for beta-sheet formation, tertiary interactions are sufficient to stabilize the correct tertiary and secondary folds in this alpha-helical protein - although this native-fold is less stable in the computational model than it is in reality. I propose to use the same parallel tempering protocols to simulated a thermo-stable variant of the WW domain, a two-stranded beta sheet with 34 residues. The objective is to demonstrate that a current force-field with explicit solvent is capable of folding beta and alpha proteins. This data will serve as a starting point for investigating the sensitivity of the melting curves and kinetics of small proteins to force-field torsional adjustments, and the basis for force-field adjustments to tune the local backbone propensity for folding studies. Initial work will focus on feasability studies. The PI expects the WW domain relaxation to be similar to that of the trpcage protein. Complete equilibration in that protein required approximately 50ns of simulation per replica, which is equivalent to about 70000 hours of calculation on a recently built intel based cluster. Initial folding events, however, were observed in only about 10ns of time. Consequently, the PI expects that 30000 hours will be sufficient to estimate a total simulation time and determine feasibility. Validation and adjustment of existing force-fields will have wide application in all areas of biological simulation. The use of empirical protein folding and denaturation to tune potentials is a relatively recent idea that can complement more detailed quantum mechanical simulations in parameterization.
Showing the most recent 10 out of 292 publications
