Coming from small molecule crystallography where this problem simply doesn't exit, I found it very difficult to understand all the fuss! The small molecule crystallographers specify the symmetry operations rather than the space group so there is nothing special about using different settings. If a program insists on starting from the space group name or number, a simple solution is to treat 'R32' and 'H32' as synonymous and to check the unit-cell dimensions to see which of the two settings it is. This requires one extra line of computer code and eliminates the problem. I know that many programs do it this way already, but unfortunately there are still a few black sheep.
George Prof. George M. Sheldrick FRS Dept. Structural Chemistry, University of Goettingen, Tammannstr. 4, D37077 Goettingen, Germany Tel. +49-551-39-3021 or -3068 Fax. +49-551-39-22582 On Thu, 12 Aug 2010, Phil Evans wrote: > As far as I know all the programs we use will handle the non-primitive > hexagonal setting of rhombohedral space groups (ie a=b != c, alpha=beta=90, > gamma =120), often denoted "H3" or "H32" (aka R3:H in cctbx). Many of them > will probably handle the primitive rhombohedral setting (a=b=c, > alpha=beta=gamma), but this is much less commonly used in macromolecular > crystallography, so some programs may fail to do the right thing (though they > should). A space group denoted "R3" or "R32" may be in either setting, > depending on the cell dimensions. I don't know what HKL2000 does, but "H32" > is not primitive, it is the centred hexagonal setting. > > The paper cited by Ivan presumably used the hexagonal setting, since they had > c != a or b > > confusing I know > Phil > > > > On 12 Aug 2010, at 00:42, xaravich ivan wrote: > > > Hello Ray, > > You can contact the author of this paper. It seems they used denzo scalpack > > for R32. > > Ivan > > > > On Wed, Aug 11, 2010 at 8:53 AM, Eleanor Dodson <c...@ysbl.york.ac.uk> > > wrote: > > If you cant import a file with symmetry labelled R32 but with a=b and > > alpha=beta = 90, and gamma = 120 then something is wrong - the symmetry > > checking looks at both name and cell.. or at least it used to.. > > > > The output mtz would have symmetry name changed to H32, but that is just a > > name - in each case the sym ops and cell should be h32 style.. > > Eleanor > > > > > > > > > > Ray Brown wrote: > > Hi, I have read all of the past messages about the confusion of H32 and R32 > > and I want to get the correct indexing sp that I can import the .sca file > > into CCP4.. > > > > I use DENZO and SCALEPACK in HKL2000. If you read the manual then it states > > that space group R32 should only be autoindexed by DENZO on a primitive > > lattice using the DENZO keyword H32. This indeed gives a primitive unit > > cell and not the hexagonal setting apparently demanded by CCP4 programs. > > > > What do you do next? It is possible in SCALEPACK to reindex the H32 > > primitive data into a rhombohedral hexagonal setting which SCALEPACK calls > > R32. Or course you cannot import this .sca file into CCP4. Why not? > > I can go into the .sca file and edit the header and change the R32 label to > > H32. Does this screw up the indexing? > > > > If anybody has successfully used DENZO/SCALEPACK/CCP4 in space group R32 > > then please tell me how to do it correctly. > > > > Cheers. > > > > Ray > > email ray-br...@att.net > > > > <am0110-1.pdf> > >
