Hello Yadong, This is what I would do in your position.
1) Integrate everything in MOSFLM 2) Enforce consistent indexing using POINTLESS (unless I am mistaken, there are alternative origin(s) in p321) http://www.ccp4.ac.uk/dist/html/alternate_origins.html ftp://ftp.ccp4.ac.uk/ccp4/6.0.2/prerelease/pointless.html 3) Bundle ALL integrated datasets into one .mtz file (being such to make sure all the batch number do not conflict)(POINTLESS may do this for you now) 4) Push through SCALA/Truncate in one giant run - selecting your 'best' dataset as a reference. Scala should use all the reflections you give it, not just the overlapping ones The rejection of redundant reflections should only really be done if you have a very good reason - redundancy (hopefully) adds to the quality of your data Scala/Truncate will give you masses of stats for individual datasets, and how well they scale together. At the end you should get a single .mtz file containing all your reflections scaled together and converted to Fs. Hope this helps, David -- ============================ David C. Briggs PhD Father & Crystallographer http://www.dbriggs.talktalk.net AIM ID: dbassophile ============================
