On 11/18/2009 05:46 AM, Steve Seibold wrote: > Having successfully install the newest version of Modeller on my new > machine (I ran some test cases), I would like to attempt the > construction of a mutant of my protein using the tutorial: > http://salilab.org/modeller/wiki/Mutate%20model. However, my protein has > metal and DNA with it and, so, I am not sure how Modeller will deal with > this. Can you inform me? I am worried that Modeller will simple fail > with the error that it can not recognize certain atoms. If it does, is > there away I can get around this? Could I do only a local (around the > mutated site) minimization only, ignoring the rest of the protein?

The easiest way to find out is to run it and see. ;) No, it won't have any problem with any HETATM residues in your protein (as long as you don't try to mutate those residues, of course). Just as for regular homology modeling, the HETATM residues will be simply copied from input model to output.

As for "local minimization", that's exactly what that script does. Only the residue you mutate is moved - every other residue in the protein remains fixed. You can, of course, change the script if you want to do something different - e.g. to allow nearby residues to move as well.

Ben Webb, Modeller Caretaker