PS. You will still need R-free flags to use this program.  I have no idea if the XPREP output includes these, or whether the =hklf3 hack will still allow them to be detected, but if not, you will need to specify those as a separate file:

r_free_flags.file_name=flags.mtz

-Nat

On Thu, Aug 22, 2013 at 9:01 AM, Nathaniel Echols <nechols@lbl.gov> wrote:

On Thu, Aug 22, 2013 at 8:43 AM, Morten Grøftehauge <mortengroftehauge.work@gmail.com> wrote:

So first the missing manual. If I google it I get this page http://www.phenix-online.org/documentation/find_peaks_holes.htm which is for the GUI. When I go back to the documentation http://www.phenix-online.org/documentation/phenix_documentation.html I can't find the link for it (which is also why I was using Google to find it in the first place). I also checked the source to see if the link was somehow misnamed.

The manual isn't "missing", I simply don't have time to cover every single advanced feature in every section, and the behavior is fundamentally the same as at least a dozen other programs (with respect to SHELX files, anyway).  So far you're the first person to ask about command-line use.

Anyway, running it from the command line with default settings is not hard.

phenix.find_peaks_holes model.pdb reflections.hkl

but I run into this error message 

Sorry: Unresolved amplitude/intensity ambiguity: /path/path/path/reflections.hkl

  SHELX reflection files may contain amplitudes or intensities.

  Please append   =amplitudes

             or   =hklf3

             or   =intensities

             or   =hklf4

  to the file name argument or parameter to resolve the ambiguity.

 and I don't how to do that. If I add .hklf3 at the end of the file name I get the same error. If I go reflections.hkl=hklf3 I just get a message that I need to supply a reflection file. 

What is the syntax?

phenix.find_peaks_holes model.pdb xray_data.file_name=reflections.hkl=hklf3

The same applies to most other programs which produce a similar error message.  This is a little confusing, I agree.  I think it's mostly a historical issue of the order in which various components of Phenix were written.  Thinking about it a little bit, I think I can hack the code to handle just "reflections.hkl=hklf3" as a special case.

I still find it inexplicable that in 2013 there are programs which are still restricted to outputting a format which is fundamentally broken when so many others are available.  We try to support any input format in Phenix, but there are limits to how much time we can invest in compensating for this.

-Nat

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