Hi Sam,

phenix.refine outputs .geo file that lists all the geometry restraints used in refinement for all atoms. You can find the atoms you are interested in and see what are the restraints targets (ideal values for bonds, angles, etc...), then compare those with the CIF file to make sure which one is used. I'm almost sure that will answer your question.

Pavel.

On 6/25/10 11:50 AM, Sam Stampfer wrote:

Does anyone know what file phenix uses to do NAG restraints when the molecule being refined has NAG groups? My groups are called "NAG". I found data_NAG.cif and NAG-B-D.cif in the phenix folder but can't tell whether it's using either of those to refine my NAGs (they don't show up in the log file).
Thanks,
-Sam

On Thu, Jun 24, 2010 at 11:42 AM, Sam Stampfer <[email protected]> wrote:

Hi,

I'm in the process of preparing 4 structures for submission to rcsb, and the validation server has pointed out numerous issues with the N-acetyl-glucosamine moieties on my structures. While I have fixed the torsion angles accordingly (using CARP from http://www.glycosciences.de/tools/carp/ as a validation checker) and they stay fixed after phenix refinement, phenix keeps generating 2-2.2 Angstrom clashes between ND2 on asparagine with C2 on the NAG (the linkage is ND2-C1 and the usual distances are at least 2.28 Angstroms). Occasionally, the ND2 to C2 distance is as low as 1.8 Angstroms. This is happening in all my structures and happens independently of the setting of wxc_scale (even for wxc_scale=0.05!) My NAG groups fit the density well, both pre- and post- refinement.

Any suggestions? Will rcsb care? These clashes are present in 12 out of 39 NAG moieties.

Thanks!

-Sam
_______________________________________________
phenixbb mailing list
[email protected]
http://phenix-online.org/mailman/listinfo/phenixbb
  