*** For details on how to be removed from this list visit the ***
*** CCP4 home page http://www.ccp4.ac.uk ***
On Tuesday 31 January 2006 10:46 am, Prem kumar wrote:
>
> I am refining atomic resolution (0.9 A) structure. I am facing
> problem with R-factor 15.2 and R-free 18.5 even after anisotropic
> refinement (including hydrogen), which is high for atomic resolution
> data. the model seems to be fine and almost every residues and water
> fitted well in electron density.
What is the R-factor for the same data + model if refined at, say,
1.1A, 1.3A, 1.5A?
> i found few alternate conformation and included them for refinement,
> but still R-factor is not coming down.
It may well be that you need to model *many* alternate conformations
at this resolution.
> data (synchrotron) was
> collected in two sets, one for high resolution and one for low
> resolution. Rsym - 5.0(55.0), I/sig I - 17.2(2.0), completness -
> 99.8(100).
You could try assigning reflections in the high-res dataset and the
low-res dataset to different scaling groups, and let the relative
scale float during refinement.
> what are the other factors I need to look into?
> Please suggest me how to proceed further.
Try to pin down whether the data is truly good all the way to 0.9A
resolution. Pay serious attention to any hydrogen geometry clashes
reported by shelx or by MolProbity. Keep looking for alternate
conformations. If you have not modeled at least 2 conformers for
each Met residue, you're not looking hard enough :-)
Have a serious look at your ADP restraints - run the structure through
the Parvati server
http://www.bmsc.washington.edu/parvati/parvati.html
--
Ethan A Merritt
Biomolecular Structure Center
University of Washington, Seattle WA
