IBM and NeSC workshop on Protein Science
National e-Science Centre, Edinburgh, March
15-16 2002
|
PROTSCAPE: An E-Science Application for AbInitioProtein Folding using a Physics-based Potential Model D.J. Osguthorpe University of Bath, Claverton Down, Bath, BA2 7AY, UK. Over the past few years I have been developing a potential model for protein folding based on the physics of the interactions involved. This is still the only model being used at CASP which does not use statistical potentials directly and all structure is generated by the potentials. The most important potentials are those that deal with the effects of the solvent environment. An underlying goal of this model, unlike most other protein folding potentials, was a correct treatment of the electrostatic features of the protein and its environment. For CASP4 a much simpler and cleaner model of electrostatic interactions (which also includes backbone hydrogen bonds) was introduced, the differential dielectric model, although it is based on previous ideas. It allows charges to have a self-energy term to correctly reproduce their exposed nature. It also allows the strength of the backbone hydrogen bonding potential to be varied with the degree of burial of the backbone. |
|
||||||||||
|
|||||||||||
