[phenixbb] Specify space group with phenix.map_to_structure_factors?

Pavel Afonine pafonine at lbl.gov
Mon Feb 15 21:16:05 PST 2016

Hi Oliver,

> Hi, I frequently use phenix.map_to_structure_factors to generate mtz 
> files from EM maps for use in Coot - it is by far the best tool I have 
> found for this purpose.
> For convenience (so I can easily build and refine one protomer of a 
> four-fold symmetric entity), I often convert these mtz files from the 
> default P1 to P4 spacegroup using CCP4, and I wonder if  it would be 
> possible to include an option to specify the output spacegroup 
> directly in map_to_structure_factors (or perhaps it is there and I 
> have missed it?)?

structure factors calculated from input map inherit crystal symmetry 
information from it. Input map is always assumed to be in P1 box. This 
prompted me to check this and I realized there was no appropriate 
assertion in the code, so I just added it: from now on the program will 
stop if the input map is not in P1.

I can add two keywords that you can set for example like this:

unit_cell=10,20,30, 90,90,90

and that will be used (if defined) to calculate structure factors map in P1.

Is that what you are asking?

> Even better would be if there were a way to check the symmetry and 
> refine the orientation of the input volume such that the map is 
> centered in the unit cell and the symmetry axis of the volume is 
> appropriately aligned - often I have aligned the input volume to 
> another reference volume in Chimera, and if the alignment is not 
> perfect the symmetry axis of the volume might be slightly misaligned 
> from what would become the fourfold axis of the output.

This sounds like a non-trivial project to me. I don't have ready to use 
tools to do this so I would have to come up with something from 
scratch.. I afraid I can't afford doing this right now.


More information about the phenixbb mailing list