Hi Charlie, expanding to P1 can be done with CCP4 sftools (see https://www.ccp4.ac.uk/html/sftools.html#expand), or with phenix.reflection_file_converter --expand_to_p1 input.mtz --mtz=output.mtz (Maybe there's yet another CCP4 tool)
Unfortunately, expanding removes the information about the differences of the molecules that are pseudo-symmetry related. I'd rather keep this information. Because scaling needs multipicity (which typically is poor in P1), I therefore suggest to scale in C2, but to merge in P1. 1. scaling in C2: If you use XDS, you obtain XDS_ASCII.HKL scaled in C2 but since it is unmerged, its reflections retain the full P1 information. If you want to be more certain that large real differences between pseudo-symmetry related reflections are not considered as outliers in scaling, increase WFAC1 to (say) 2. 2. create XDSCONV.INP with SPACE_GROUP_NUMBER= 1, but the UNIT_CELL_PARAMETERS of the C2 cell 3. run XDSCONV to obtain a merged MTZ file in P1, with the C2 cell 4. run CCP4 reindex to obtain the conventional P1 cell (sorry that this is so cumbersome due to the reindexing) It is clear that the statistics you obtain from XDS in step 1 do not apply to the P1 data you obtain in step 3 and 4, but the data should be better than the expanded data. (And the expanded data have the same problem concerning their statistics) Hope this helps, Kay On Thu, 3 Sep 2026 11:22:35 +0000, Nichols, Charlie <[email protected]> wrote: >Hi, > >I have a very peculiar crystal structure which has been challenging to solve: > >Prot-X + VHH – nominally expected to be a 1:1 complex > >Data auto-process as C2 >CRYST1 132.084 56.245 86.262 90.00 108.40 90.00 C 1 2 1 > >2 copies of Prot-X easily located by molecular replacement, Phaser fails to >find the VHH > >Manual assessment of density in crystal voids reveals a density patch about >the right size / shape for a VHH. >Manual placement of a VHH into this density shows it lies across the C2 2-fold >so the lattice symmetry places an inverted copy of the placed VHH in the same >location. > >Expanding the symmetry to P1 gives 4-chains, arranged in a square ring, with a >hole in the middle where the VHH sits, sterically only 1 VHH can sit in the >hole >CRYST1 56.245 71.780 86.262 106.88 90.00 113.07 P 1 > >I overlaid a computational model of Prot-X + VHH onto each of the 4 chains, 2 >positions align reasonably with the density (related by 2-fold as expected), >the other 2 do not. >If I refine either of these models in P1 with twin refinement then I get a >reasonable map, especially where the VHH contacts Prot-X as the alternative >2-fold related position does not overlap at the contact surface. > >Some parts of the density look really good, some are a bit confused / look >like some rebuilding is required, though this may be largely residual density >from the alternate orientation. > >Data resolution is modest with fairly noisy outer shell data so I can process >to 2.6A with much better stats in C2 but only 2.8A and significantly worse >stats in P1. > >I have had similar cases before in monoclinic/orthorhombic and >trigonal/hexagonal and was able to obtain better maps by symmetry expanding >the higher resolution processing from the higher symmetry spacegroup to the >lower symmetry spacegroup rather than processing in the lower symmetry. > >I would therefore like to take the C2 processed data, symmetry expand to P1 >and then use these expanded data for twin refinement to try and improve map >quality. > >I am however not sure how to expand / transform the data from the C2 lattice >to the P1 case. > >Help appreciated… > >Thanks, Charlie. > >######################################################################## > >To unsubscribe from the CCP4BB list, click the following link: >https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB&A=1 > >This message was issued to members of www.jiscmail.ac.uk/CCP4BB, a mailing >list hosted by www.jiscmail.ac.uk, terms & conditions are available at >https://www.jiscmail.ac.uk/policyandsecurity/ ######################################################################## To unsubscribe from the CCP4BB list, click the following link: https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB&A=1 This message was issued to members of www.jiscmail.ac.uk/CCP4BB, a mailing list hosted by www.jiscmail.ac.uk, terms & conditions are available at https://www.jiscmail.ac.uk/policyandsecurity/
