[phenixbb] Refining Anomalous Groups

[Terwilliger, Thomas C]

Hi Brennan,

I'm sorry, part of the problem is that autosol exptl_fobs_phases_freeR_flags_1.mtz file does not retain the anomalous information.  This is now on my list be fixed (not quickly though because I will be able to do it only when I convert a whole lot of other things).

So in your case the best thing is to run phenix.refine with your original mtz file.  You can create a freeR set in the GUI.  If you do this, then use your new freeR set for everything from now on.  As Nat mentions, if you choose "F(+),SIGF(+),F(-),SIGF(-)" or something like it then it should work.

All the best,
Tom T
________________________________________
From: phenixbb-bounces at phenix-online.org [phenixbb-bounces at phenix-online.org] on behalf of Brennan Bonnet [Brennan.Bonnet at lightsource.ca]
Sent: Monday, November 19, 2012 8:03 PM
To: PHENIX user mailing list
Subject: Re: [phenixbb] Refining Anomalous Groups

Ok.  So I used the original mtz file created from my data to determine the heavy atom substructure using HySS which I then used as input for AutoSol.  Next I carried the files through to AutoBuild these files are:

exptl_fobs_phases_freeR_flags_1.mtz (experimental data)
resolve_1.mtz (initial map)
overall_best_placed.pdb (ligands, starting model)
ha_1.pdb_formatted.pdb (heavy-atom sites)

Finally I carried these through to phenix.refine and now is when it complains that my data is no longer anomalous and I can't find a file that phenix is happy with.

Everything worked smoothly until this step.  Any idea what file(s) I might be looking for?  I have tried some data from earlier steps which are unmerged and it still complains.

~Brennan~

From: phenixbb-bounces at phenix-online.org [phenixbb-bounces at phenix-online.org] On Behalf Of Nathaniel Echols [nechols at lbl.gov]
Sent: Monday, November 19, 2012 8:29 PM
To: PHENIX user mailing list
Subject: Re: [phenixbb] Refining Anomalous Groups

On Mon, Nov 19, 2012 at 6:11 PM, Brennan Bonnet <Brennan.Bonnet at lightsource.ca<mailto:Brennan.Bonnet at lightsource.ca>> wrote:
I am trying to use phenix.refine to refine partially occupied xenon sites using the Anomalous Groups selection under "Refinement settings".  My list of anomalous scatters initially consisted of 22 peaks as located by HySS.  Some of these were removed as they originated from S atoms from Met and Cys residues and a few others were removed after previous refinement steps due to having zero occupancy.  My remaining sites consist of my xenons (hopefully) and other metals such as Mg or Zn which are mixed in and look like partially occupied xenon sites.  My problem is that when I try to define the 11 remaining sites for the "refine anomalous groups" selection the software complains:

Invalid combination: F-obs not anomalous, but refinement of 11 f_double_prime values requested (11 scatterers): Please provide anomalous reflection data or fix all f_double_prime values (see refinement.refine.anomalous_scatterer.group input blocks)

I can't figure out exactly what I need to change to make it work and can't find a guide anywhere.  Do I need to provide additional data or is the syntax for the sites incorrect maybe?

No, the error message means exactly what it says: you need to supply anomalous data as input.  This almost always means that the selected data labels should have "+" and "-" in them, and there will usually be 4 labels (e.g. "F(+),SIGF(+),F(-),SIGF(-)").  Data named "F,SIGF", on the other hand, is almost always merged/non-anomalous (although I think CNS files may break this assumption).

-Nat
_______________________________________________
phenixbb mailing list
phenixbb at phenix-online.org
http://phenix-online.org/mailman/listinfo/phenixbb