On 10/05/2009 10:12 AM, Liam Browne wrote: > I have a question about the most effective way of modelling a protein > that is in a different conformation to the structure? I am hoping to > make a homology model, then rotate a helix about its center and allow > modeller to remake the model (allow energy minimisation etc) to > determine the effects on the whole protein. Is there a way to do this?
You can certainly build homology models (of course), and Modeller includes several methods for rotating, translating or otherwise transforming a set of atoms: see http://salilab.org/modeller/9v7/manual/node233.html for an example. The only problem you will run into is the minimization, since you will have to change the energy function somehow. The default energy function tries to make the model look like the template, and this is obviously not going to work if you want to build the model in a different conformation. You may be able to use a conventional dynamics package such as CHARMM or GROMACS though to relax the structure.
Ben Webb, Modeller Caretaker