Hi I have been using a SePeak.hkl file in my refinement (CNS, and then switched to Phenix). This was created from SePeak.sca to SePeak.mtz and then to SePeak.hkl with ccp4i's (scalepack2mtz and mtz2various). I want to try using the group anomalous command to see is my Rfree decrease with the inclusion of the anomolous data. The problem I am encountering is that the SePeak.hkl contains no anomolous inforomation. If I make a new mtz (I have used reflection_file_converter to create a new SePeak#2.mtz from my original SePeak.sca and this does contain + and - columns) I can't find a way to merge the test set from the SePeak.hkl file. And generating a new test set raises othere issue that have been discussed in this bb before. Does this make sense? And is there a way to merge a test set from one file and a F or I obs. Mark