On Sun, Jan 13, 2013 at 11:12 AM, Xiang Li
1 I worked on a DNA structure and wanted to add restraints during the refinement. I used GUI version and chose "secondary structure restraints"->"restraint base pairs". However, there were some distortion of the base pairing after refinement. Is there a way to assign the base pairing between specific base pairs and force the base pairings in the GUI version?
Sort of - right now you have to generate a parameter file (I will see about fixing this in the future). I would start by running this command in the terminal: phenix.secondary_structure_restraints model.pdb which should generate parameters for using in phenix.refine. You can edit these as you see fit, or add new ones following the same syntax. (Note that you shouldn't normally have to specify the base pair type as this will be guessed automatically, but there is always a chance that it will guess wrong if the input model is poor.) If you're willing to share them (off-list!), we would be interested in seeing the input and output files from the refinement that went poorly, as we would like for these restraints to be as automatic (and harmless) as possible.
2 I want to compare two DNA structures to see the RMSD. Each of them were composed of 4 strands. Seems "structure comparison" software can not work with such multiple strands DNA structures. Do you have any suggestions on how to do the comparison please?
Sorry, the structure comparison program is very specific to proteins. For something like this I would normally just use the superpose functions in Coot and/or PyMOL. -Nat