Another idea is to use qfit:

http://smb.slac.stanford.edu/qFitServer/qFit.jsp



On Mon, Feb 13, 2012 at 3:58 PM, Nathaniel Echols <nechols@lbl.gov> wrote:
On Mon, Feb 13, 2012 at 11:53 AM, Subhani Bandara <ramssb17@gmail.com> wrote:
> I have a protein where three consecutive amino acids have alternative
> conformations. I edited the PDB file to have those altloc and made total
> occupancy of A and B to 1.00. But after refinement both A and B are
> overlapping and I could see the positive density for one conformation away
> from that (may be b'se I didn't change coordinates of B conformation). How
> can I move the B altloc in to the positive density. Do I have to edit the
> coordinates on PDB and if so how can Ido that?

phenix.refine won't automatically move the alternate conformations for
you - you need to do this with Coot.  An alternative is to try
simulated annealing to try to "shake" them into the correct place, but
rebuilding in Coot is probably going to be faster and more accurate.

-Nat
_______________________________________________
phenixbb mailing list
phenixbb@phenix-online.org
http://phenix-online.org/mailman/listinfo/phenixbb