It’s probably better to combine the MTZ files with Pointless rather than sortmtz as that will also check the point group Phil
Sent from my iPhone > On 20 Oct 2017, at 18:57, Gottfried Palm <p...@uni-greifswald.de> wrote: > > Dear all, > technically, the ccp4bb (especially Conn Mallett and Jim Pflugrath) helped > to resolve the problem: The procedure to get ccp4-readable data from the .ref > files uses dtreflnmerge (provided as part of CrystalClear / d*trek from > Rigaku) from the command line, for example: > > dtreflnmerge.exe dtprofit.ref dtprofit_ascii.ref -ascii > > These .ref files could be converted by dtrek2scala to an mtz file in ccp4. > These mtz files (one for every scan) were merged with sortmtz. > The sorted multibatch mtz file was scaled with aimless. > > The structure was not solvable, though, which is a bit puzzling to me, > because the diffraction pattern is clean to better than 1 A resolution, the > unit cell only ca. 1000 cubic Angstrom (small molecule) and the crystals > contain Pd with an organic ligand. I am thus still wondering, if something > gets messed up in this procedure. I have also tried processing with xds, > mosflm and xia2 dials, they failed at indexing. I don't have the resources to > figure this out soon, but anybody, who is curious to try the processing is > welcome. Contact me directly in December, success will be duly acknowledged. > > Greetings > Gottfried > > original post: > > Problem with ref format (Rigaku) > > Dear all, > > this is a question about scaling data integrated in CrystalClear (Rigaku > data processing gui based on d*trek) in ccp4. > Since scaling dtprofit.ref files from different scans is sometimes poor or > even failing within CrystalClear (i.e. with dtscaleaverage after merging > them), I used to try scaling them with scala. I am facing this problem mainly > with high resolution / small molecule data collection, where I need up to 20 > scans, each 90-180 degrees, for complete low and high resolution. > > The procedure, that worked, was using > > dtrek2scala for each scan (with scan1.ref and output_scan1.head, then a > second run of dtrek2scala for scan2.ref and output_scan2.head, etc.) to > create scan1.mtz, scan2.mtz, etc. > sortmtz with scan1.mtz, scan2.mtz, ... to create a multibatch mtz file > scala with the multibatch.mtz file to create the final scaled mtz file > > Some time ago Rigaku changed the format of the .ref files, so dtrek2scala is > not working any more. > Is there a possibility to change the new .ref format to the old one? Or can I > read the new .ref files in scala or better aimless directly? > > The alternative to process the images in xds hits a similar problem: The > format of the images has also changed and the new .img files (from the > Saturn92 detector) are not read anymore (whereas they used to be processable > in xds before). > > Greetings > Gottfried > > > Dr. Gottfried Palm > Ernst-Moritz-Arndt-UniversitÀt > Inst. für Biochemie (MNF) > Abt. Biochemie I > Felix-Hausdorff-StraÃe 4 > 17489 Greifswald