Everyone,

The problem seems to relate to using Real Space Refinement in phenix.refine, which at the final stages of refinement I usually turn off, but did not this time.  Turning off XYZ (real-space) in the GUI allows WXC to work properly.

This does seem to indicate a bug in the WXC optimization routine. Years ago I implemented an optimization scheme in CNS based on target Bond-RMSD values which can be roughly related to the number of observations / number of atoms ~ (#reflections>5sigma)/(#atoms).  The target can then be limited to physically reasonable values 0.004 < Bond-RMSD < 0.025, and you never see a bond RMSD of 0.035 Å.

Best regards,
Mark

Mark Andrew White, Ph.D.
Associate Professor of Biochemistry & Molecular Biology,
Manager, SCSB Macromolecular X-ray Laboratory
UTMB, Galveston, TX
http://xray.utmb.edu

From: Dorothee Liebschner <dcliebschner@lbl.gov>
Sent: Monday, August 10, 2020 10:11 PM
To: Nigel Moriarty <nwmoriarty@lbl.gov>
Cc: White, Mark <mawhite@UTMB.EDU>; phenixbb@phenix-online.org <phenixbb@phenix-online.org>
Subject: Re: [phenixbb] Phenix.Refine fix WXC has no effect
 
WARNING: This email originated from outside of UTMB's email system. Do not click links or open attachments unless you recognize the sender and know the content is safe.

Hi Mark,

While waiting for Pavel to return, you could do the following:
Check if the rmsd goes up because all bond rmsds increase or because there are a few outrageous outliers. The validation tab after refinement lists the worst offenders for geometry restraints. 

Best wishes,

Dorothee

On Mon, Aug 10, 2020 at 3:12 PM Nigel Moriarty <nwmoriarty@lbl.gov> wrote:
Mark

Pavel is best to answer this but he will return on Thursday. Someone else may offer guidance until then.

Cheers

Nigel

---
Nigel W. Moriarty
Building 33R0349, Molecular Biophysics and Integrated Bioimaging
Lawrence Berkeley National Laboratory
Berkeley, CA 94720-8235
Phone : 510-486-5709     Email : NWMoriarty@LBL.gov
Fax   : 510-486-5909       Web  : CCI.LBL.gov


On Mon, Aug 10, 2020 at 1:10 PM White, Mark <mawhite@utmb.edu> wrote:
Hello,

I am refining a 3.5 Å structure using the latest stable version of Phenix.Refine. After switching to the latest version the Bond RMSDs exploded to ~ 0.035 Å, when using optimize wxc.  Turning off Optimize and using the "Fix WXC" parameters in the GUI with increasingly smaller values had almost no effect on the final bond rmsds, which seems to explode from reasonable values in the "3_xyz" step of refinement.  Am I missing something?

Best regards,
Mark

Mark Andrew White, Ph.D.
Associate Professor of Biochemistry & Molecular Biology,
Manager, SCSB Macromolecular X-ray Laboratory
UTMB, Galveston, TX
http://xray.utmb.edu
_______________________________________________
phenixbb mailing list
phenixbb@phenix-online.org
http://phenix-online.org/mailman/listinfo/phenixbb
Unsubscribe: phenixbb-leave@phenix-online.org
_______________________________________________
phenixbb mailing list
phenixbb@phenix-online.org
http://phenix-online.org/mailman/listinfo/phenixbb
Unsubscribe: phenixbb-leave@phenix-online.org


--
Project Scientist, Molecular Biophysics and Integrated Bioimaging
Lawrence Berkeley National Laboratory
1 Cyclotron Road, M/S 33R0345
Berkeley, CA 94720
Tel: (510) 486-5709
Fax: (510) 486-5909