I should do more digging, but I hope maybe there is a simple explanation and someone has seen this before. On some datasets (collected at SSRL) I get SCALA reporting average mosaicity of 0.0. This probably happens at the integration stage, and for this whole set of datasets *always* happens when I use the autoxds scripts. When I go with mosflm/scala, it still happens for some, but not all datasets. I can process those that fail mosflm using denzo/scalepack, but it takes a bit of tinkering with parameters (diffraction is admittedly messy).
Interestingly, it seems that at least in some cases all the other SCALA statistics are perfectly fine. I haven't checked yet how these will behave in refinement, but I suspect it will look OK too. I have found this by googling http://www.mail-archive.com/ccp4bb@dl.ac.uk/msg00422.html but it's from 2005 and I wonder if things changed since. Andrew mentions the multiple close lattices as one of the possible reasons, and it is indeed fairly common for these datasets. I cannot find anything in SCALA manual about mosaicity refinement, so I assume that scala (unlike scalepack) does not do that. So if I am to overcome the zero mosaicity issue by fixing it at mosflm stage, how important it is to get it close to the actual value? Or is it enough to just keep it sufficiently high to prevent rejections of legit spots? And, if I may ask one last question, is there a way to fix mosaicity in imosflm gui (I can *fix* it, but doesn't seem to be possible to choose a specific value). Cheers, Ed. -- After much deep and profound brain things inside my head, I have decided to thank you for bringing peace to our home. Julian, King of Lemurs
