at 1.2A resolution, you can probably go higher than 2 with the weighting term,
especially at the end of your refinement, when the model is near
completeness. You can try 'weight auto' to let refmac5 figure out the correct
weight (you have to use 'Run&View Com file' to edit the script manually to do
A large value for the weighting term means that the crystallographic data are
weighted more strongly than the geometric restraints, hence you probably want
to increase rather than decrease that value.
On Friday 18 May 2007 11:13, XJ XJ wrote:
> Dear all,
> I am having a problem with refining my structure using
> refmac5. Basically, I am trying to make some small
> changes to several places on the backbone. But the
> thing is that, after restrained refinement, the
> changes I made would be all gone and the chain stayed
> exactly where they were before I made those changes.
> I'm using Turbo for building the model. The
> resolution I'm working on now is 1.2 A.
> The refmac was working fine when I was working at a
> lower resolution (2 A) and I was able to rebuild a
> whole helix and refine successfully with refmac.
> Right now, I am still able to change rotamers of
> single residues and refmac will keep these type of
> changes for me. It is only when changes are made on
> the backbones that refmac will refine the structure
> back to what it was before the changes were made.
> Other than refining to highest resolution 1.2 A, I'm
> pretty much using all the default parameters from the
> restrained refinement, except that, under Geometric
> parameters, I changed all the overall wt restraints
> from 1.0 to 2.0. Weighing term has been varied from
> 0.3 to 0.5 and no difference was made.
> I guess it's probably some small mistakes I made while
> running the refmac, but I can't figure out what it is.
> Can anyone help me out real quick? Thanks!
>_________Boardwalk for $500? In 2007? Ha! Play Monopoly Here and Now (it's
> updated for today's economy) at Yahoo! Games.
800 Blackburn Road
Clayton, VIC 3168