*** For details on how to be removed from this list visit the *** *** CCP4 home page http://www.ccp4.ac.uk ***
While waiting for and trying the other more intelligent suggestions, you might also try letting arp/warp have a go at your stuck solution. If in fact the space group is correct and you MR solution has some validity but is not quite good enough to make rebuilding easy, A/W might kick in and build your model for you. I am playing with such a case: MR from a distant homolog, with no attempt to to taylor the model to be more like the target sequence. After refinement in CNS stalls at R/Rfree 0.43/0.44 (no manual rebuilding), A/W builds (in one example) 1818 residues and docks 1530 in sequence (out of ~2250 total). Using resolution 9 to 1.9 A. In a way A/W's approach would be taking the suggestion to revert to polyalanine (in case of permuted segments) one step farther- revert to free atoms. I use the CCP4i gui. Select Model Building, arp/warp, task "automated building starting from existing model", give it your current model and an mtz file with Fobs and SigFobs, and let it run. Ed
