hi (1) provided the predictions match the spots (i.e. indexing etc has been successful) the best things to look at in the Integration task initially are the graphs - if they vary reasonably smoothly, and there are no big jumps then things have probably gone okay.
(2) Really, the only way to reduce the mosaicity is to grow better crystals. Are you getting much larger values than with hkl2000? How many overlaps are you getting (both as the overall number and as a fraction of the whole?); there may be nothing wrong at all... (3) For everything apart from finding the heavy atom sub-structure with SHELXC/D/E, you want the MTZ file from ctruncate (so this is the one you want for molecular replacement). For SHELXC/D/E, you really want unmerged F^2 values which can be obtained by using the Aimless option "output polish unmerged" (as a processing option under "Integration"). HTH Harry -- Dr Harry Powell, MRC Laboratory of Molecular Biology, Francis Crick Avenue, Cambridge Biomedical Campus, Cambridge, CB2 0QH Chairman of European Crystallographic Association SIG9 (Crystallographic Computing) > On 26 Aug 2014, at 16:51, 陈昂 <[email protected]> wrote: > > Dear all： > > It is my first time to use IMOSFLM instead of HKL2000. Here are my problems. > Hope you can help me. > > 1、what parameters can indicate the result quality of my imosflm? mosaicity > ,anything else? > > 2、what can I do to reduce the mosaicity and decrease bad spots,resolution or > the spot finding range? > > 3、after imosflm, I've got some MTZ maps such as original one and pointless > one,aimless one, ctruncate one. Which one should I take to the next step?what > steps do I need to take before molecular replacement and why? > > > THANKS a lot > > > Peter Chen
