Old way of Space group determination: Difference between revisions

Space group determination can be done within [[XDS]] by following these steps:

* an initial data reduction in P1, using SPACE_GROUP_NUMBER=0 (or even omitting that line)

* after CORRECT has thus run in P1, one may try other spacegroups by inspecting the table '''DETERMINATION OF LATTICE CHARACTER AND BRAVAIS LATTICE''' in [[IDXREF.LP]] (see the example below). The [[jiffies|sortlattices]] jiffy is useful here.

* for each possible lattice character, starting with the one corresponding to highest symmetry, do:

** look up space groups corresponding to lattice characters from the list '''BRAVAIS-TYPE / POSSIBLE SPACE-GROUPS FOR PROTEIN CRYSTALS''' in [[IDXREF.LP]] (repeated at bottom of this article).  

** modify SPACE_GROUP_NUMBER=<number according to that list>

** modify or insert REIDX=<12 numbers corresponding to the lattice character, from the table>

** modify or insert UNIT_CELL_PARAMETERS= according to table, but make them obey space group requirements (e.g. orthorhombic: all angles 90°, tetragonal and trigonal: a=b, and so on).

** run XDS with JOB=CORRECT. Inspect [[CORRECT.LP]] and note R-factors, I/sigma and a and b modifiers of standard deviations

** for many Bravais types (defined at [http://www.iucr.org/iucr-top/cif/cifdic_html/2/cif_sym.dic/Ispace_group.Bravais_type.html]), there are several possible point groups (tp and tI: 4, 422; hP: 3, 6, 312, 321, 622; hR: 3, 32; cP, cF and cI: 23 and 432) and thus several possible space groups. In these cases, each point group has to be tested: the SPACE_GROUP_NUMBER has to be changed to one of those consistent with the point group, and the CORRECT step re-run.  

* repeat for each possible lattice character

* finally decide on the correct Bravais lattice and point group by comparing R-factors (in particular R_meas), and, from there, come up with possible space groups by looking at the table of systematic absences along the h,0,0 0,k,0 and 0,0,l axes of the diffraction pattern.