Depends on what you are trying to do, but if all you want to do is to move your solutions into whichever unit cell / origin / assymetric unit best matches your reference, a quick workaround is this: 1) Load your molecule in Pymol, let it generate a healthy number of symmetry mates for each chain. (Changing to a ribbon model and low quality rendering beforehand helps) 2) Then load your reference molecule into the same session. 3) Manually click through the symmetry copies to identify which ones are the best match to the reference and write those to a new file. I often do this hallway through rebuilding, and the resulting model can (normally) be used directly for refinement behaving as your original one.
================================ Jose Antonio Cuesta-Seijo, PhD Carlsberg Laboratory Gamle Carlsberg Vej 10 DK-1799 Copenhagen V Denmark Tlf +45 3327 5332 Email josea.cuesta.se...@carlsberglab.dk<mailto:josea.cuesta.se...@carlsberglab.dk> ================================ From: CCP4 bulletin board [mailto:CCP4BB@JISCMAIL.AC.UK] On Behalf Of Eleanor Dodson Sent: Thursday, September 11, 2014 11:44 AM To: CCP4BB@JISCMAIL.AC.UK Subject: [ccp4bb] making sure related molecules occupy the same space Here is a very common scenario for me. You run several MR jobs, say to search for 2 mols/asymm unit , maybe with different models or using different programs.. You get several solutions, probably on different origins and/or related by some symmetry operator. How best to overlap them? My way is this: Use PISA to assemble all solutions from each run into a compact molecule (It does this by applying allowed symmetry operators.) Calculate SFS from each solution, combine them and use PHASEMATCH to find the origin shift . Move solution two to the same origin as solution one. Now check for fitting - remembering that if both solutions have chains A B there is no guarantee that A2 maps to A1 and B2 to B1. So I superpose A2 onto A1, A2 onto B1 etc till I find a transformation matrix that is indeed a symmetry operator.. (You can do this in coot providing you can find the output matrix on the screen!) Then apply that symmetry to move A2 B2 to overlap A1 B1 ( or maybe B1 A1) as appropriate.. Obviously this is a pain in the neck! Does anyone have a neater solution?? eleanor
