First rule out the possibility that it is something coot did to you model by running the same command using the unmodified output from the first phenix.refine run. If you still get the same error, I think it is because phenix.refine disapproves of your choice of Free-R flags. On the first run you explicitly told him you want only 1000 free reflections, so he assumes you know what you are doing and goes ahead. On the second run using the free-R flags from the mtz, it could be an oversight so he wants to make sure you know what you are doing. You are required to confirm with this "disable_suitability_test=True" option. Just my guess, presumably you will get a more definitive answer soon. eab On 11/26/2014 08:52 PM, Yamei Yu wrote:

Dear all,

I use phenix.refine to refine a structure.

Firstly, I used --dryrun to generate an .eff file. The command is "phenix.refine --dry_run 7_44.mtz 7_44.pdb refinement.input.xray_data.r_free_flags.generate=True refinement.input.xray_data.r_free_flags.max_free=1000” Then I use run the program by typing “phenix.refine 7_44_refine_001.eff” It runs correct!

After that I use coot to modulate some parts of the structure, then I want to refine the structure again. I type “ phenix.refine 7_44_refine_002.def” This time I got the following error message:

Not a suitable array of R-free flags: refinement.input.xray_data.r_free_flags.label=R-free-flags

Sorry_Not_a_suitable_array: Not a suitable array of R-free flags: refinement.input.xray_data.r_free_flags.label=R-free-flags To override the suitability test define: refinement.input.xray_data.r_free_flags.disable_suitability_test=True

I checked the generated file by phenix “7_44_refine_data.mtz” I does contain a column labeled as R-free-flags. I paste part if it below. * Space group = 'P1' (number 1)

OVERALL FILE STATISTICS for resolution range 0.000 - 0.277 =======================

Col Sort Min Max Num % Mean Mean Resolution Type Column num order Missing complete abs. Low High label

1 ASC -32 31 0 100.00 -2.2 11.9 49.44 1.90 H H 2 NONE -46 45 0 100.00 -2.5 17.8 49.44 1.90 H K 3 NONE 0 49 0 100.00 18.8 18.8 49.44 1.90 H L 4 NONE 0.9 271.3 0 100.00 26.52 26.52 49.44 1.90 F F-obs 5 NONE 0.2 5.1 0 100.00 1.25 1.25 49.44 1.90 Q SIGF-obs 6 NONE 0.0 1.0 0 100.00 0.01 0.01 49.44 1.90 I R-free-flags

No. of reflections used in FILE STATISTICS 143215

Then I regenerate a new mtz file with FreeRflag by XDSCONT, then I revise the .def file accordingly and rerun the program. I got same error message:Not a suitable array of R-free flags: refinement.input.xray_data.r_free_flags.label=FreeRflag

Sorry_Not_a_suitable_array: Not a suitable array of R-free flags: refinement.input.xray_data.r_free_flags.label=FreeRflag To override the suitability test define: refinement.input.xray_data.r_free_flags.disable_suitability_test=True

I checked my new mtz file, it dose contain a column labeled as FreeRflag

Why there is an error message? How to solve it?

Thank you so much!!

Best wishes!

yamei

************************************************

Yamei Yu State Key Laboratory of Biotherapy／Collaborative Innovation Center of Biotherapy, West China Hospital, Sichuan University,Chengdu,610041, P.R.China Tel: 15882013485 Email: [email protected] mailto:[email protected] [email protected] mailto:[email protected] [email protected] mailto:[email protected]

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