[cctbxbb] Generating crystal copies

[Nicholas Pearce]

Hi Everyone,

I'm currently trying to generate crystal contacts/copies. What I want out at the end is a list of symmetry operations to generate the crystallographic copies that form contacts with my input model.

I've been using the asu_mappings object, but I'm not sure I'm using the output correctly or if it's the right thing to use.

Code snippet used to calculate the mappings:

   # Extract the xray structure from the reference hierarchy
    ref_struc = ref_hierarchy.extract_xray_structure(crystal_symmetry=crystal_symmetry)

    # Extract the mappings that will tell us the adjacent symmetry copies
    asu_mappings = ref_struc.asu_mappings(buffer_thickness=5)

    # Symmetry operations for each atom
    mappings = asu_mappings.mappings()

I printed out all of the mappings, as I presumed the vast majority of the them would be the identity, meaning that any non-identity operations would be from atoms in the buffer area I have defined. These non-identity operations would presumably be the ones that would generate the crystal contacts.

However, a large number of the atoms only had one mapping, which was not the identity - e.g. asu_mappings.get_rt_mx(x).as_xyz() might give '-x+1/2,-y+1/2,z-1/2' for an atom with only one mapping.

The effect of this is that the operations given by this generate symmetry copies that do not contact my input model in real space.

I presume this means that asu_mappings is mapping the atoms in my model to some other arbitrary asu, as well as the atoms in the buffer zone around it? Therefore the symmetry operations given are for these mapped atoms?

Can anyone help, or suggest an alternative way to generate the crystal contacts?

Thanks,
Nick
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://phenix-online.org/pipermail/cctbxbb/attachments/20150407/8f3253d8/attachment.htm>