No, this is no real problem as we have now solved it locally by deleting or rearranging columns in sftools. Using phenix.reflection_file_convert would work just as well. The point is that input and output formats tend to be obscured for the GUI users (which tend to be the younger users). Having a phenix page or wiki which briefly informs users of file handling and input/output data order would be helpful. Some of our efforts involve comparing CCP4 and Phenix results, which means that compatible mtz files are required. The nuts and bolts of CCP4 tends to be a little more transparent than in Phenix, but I like the data handling organization in Phenix. As the two program suites diverge in philosophy and implementation, the differences will need to be highlighted for users who want to use both. Thanks for the help.

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

R. Michael Garavito, Ph.D.

Professor of Biochemistry & Molecular Biology

513 Biochemistry Bldg.

Michigan State University

East Lansing, MI 48824-1319

Office: (517) 355-9724 Lab: (517) 353-9125

FAX: (517) 353-9334 Email: [email protected]

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

On Oct 21, 2011, at 6:04 PM, Ralf Grosse-Kunstleve wrote:

Hi Michael,

I did notice that CCP4 6.2.0 outputs the mtz file with this column order:

H K L FreeR_flag F_S13E_X2 SIGF_S13E_X2 DANO_S13E_X2 SIGDANO_S13E_X2
F_S13E_X2(+) SIGF_S13E_X2(+) F_S13E_X2(-) SIGF_S13E_X2(-) ISYM_S13E_X2
IMEAN_S13E_X2 SIGIMEAN_S13E_X2 I_S13E_X2(+) SIGI_S13E_X2(+) I_S13E_X2(-)
SIGI_S13E_X2(-)

This new F,SIGF,DANO,SIGDANO order explains why the four MTZ columns are combined into an anomalous array. (The phenix tools have been doing this for many years.)

Is this an actual problem for you?

You could use phenix.reflection_file_convert --non-anomalous to obtain a non-anomalous array.

In phenix.refine you can say force_anomalous_flag_to_be_equal_to=False.

Ralf

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