Re: [ccp4bb]: refinement problems

["Eleanor J. Dodson" <ccp4@ysbl.york.ac.uk>]

***  For details on how to be removed from this list visit the  ***
***          CCP4 home page http://www.ccp4.ac.uk         ***

Sean Johnson wrote:
> 
> ***  For details on how to be removed from this list visit the  ***
> ***          CCP4 home page http://www.ccp4.ac.uk         ***
> 
> I am having problems refining a structure in a new space group.  The R and
> Rfree values are stuck at 27.5 % and 31.4 %, respectively.  I have been
> using CNS for refinement, but I am open to any other suggestions.
> 
> Here are a few details...
> 
> The spacegroup is P3(1)21.  I have complete data to 1.95 A (99% in the
> highest resolution bin).  The Rsym is ~6.5%.  The structure was solved by
> molecular replacement using the same protein solved in a different space
> group.
> 
> My model contains ~600 residues.  For the most part, the density looks
> great.  However, there is a region of ~30 residues where the density
> breaks up.  The B-factors refine to about 85 A^2 in this region (the
> average overall B-factor is 45 A^2).  I have calculated composite
> simulated annealing omit maps.  I also calculated a map using the prime
> and switch routine in resolve.  None of these maps were helpful in
> retracing the model.  I have concluded that the region is simply
> disordered.  I even tried removing that region from the model and let a
> solvent correction flatten it out, but there was no movement in the R
> factors.
> 
> When I run Wilson scaling using Truncate, it tells me that I have
> anisotropic data.
> 
> Are any of these clues helpful?  I'd be happy to hear any suggestions.  I
> need to complete this structure yesterday.
> 
> Thanks,
> 
> Sean Johnson
> Duke University
> sjj1@duke.edu


 Well - ruling out the possibility that it is actually in P3221 -
I would try TLS parameterisation in REFMAC5 - that is good at handling 
domain movement.

But of course if the piece is totally disordered, no crystal analysis
will help!
Eleanor