Dear Graeme, here is what I would do, or what I would like to have.
If you are able to identify the Laue group of the data with some degree of certainty, then all of the processing and scaling should be carried out in this Laue group. Then, by looking at systematic absences you may give probabilities for each of the possible space groups, i.e. each of the eight possibilities in P-orthorhombic. Typically one option will have the highest probability and this is the one which should be written out. In a second run, the user should be given the choice of overriding this. Now, for space groups such as P222_1, this should always be reindexed to standard setting, if it turns out to be the one with the highest probability. Hope that helps, Manfred. ******************************************************************** * * * Dr. Manfred S. Weiss * * * * Team Leader * * * * EMBL Hamburg Outstation Fon: +49-40-89902-170 * * c/o DESY, Notkestr. 85 Fax: +49-40-89902-149 * * D-22603 Hamburg Email: [EMAIL PROTECTED] * * GERMANY Web: www.embl-hamburg.de/~msweiss/ * * * ******************************************************************** On Mon, 21 Jan 2008, Winter, G (Graeme) wrote: > Hi All, > > A user question about the xia2 behaviour has opened a pot of worms, and > I thought I would ask the community for opinions. If (for example) you > are using an automated data processing or analysis tool, and the > systematic absences suggest a spacegroup choice, what would you like to > do: > > (1) nothing - just mention this in the output > (2) assign the "base" version of this spacegroup (e.g. P41212 to > represent that or it's enantiomorph) > (3) create multiple copies of the reflection file with all of the > spacegroup options > > As a further question, if the spacegroup looks like P 2 21 21 (say) > would you like this to be reindexed to the standard setting? > > Now, I suspect that there will be a wide range of opinions on this. > > Following #1 will give possibly strange effects if truncate tries to > inflate systematically absent reflections > Following #2 will result in reflections being removed by truncate > #3 gives lots of reflection files and lots of mess > > Currently I follow #2 with reindexing to the standard setting. > > There have been discussions in the past of being able to flag "or > enantiomorph" in the spacegroup definition in the mtz file. This would > be useful here, but would not really help with the reindex or no > question... > > Thanks! > > Graeme > >
