Many thanks for the reply. Unless I have not RTFM properly, cphasematch only modifies one column of phases at a time, and outputs the modified phases to a new column. It also doesn't appear to take a reference PDB (corresponding to the reference mtz) as a means to calculate the appropriate rot/trans matrix for the moving PDB.
I've managed a work-around using reforigin where I first superimpose the moving PDB onto the target PDB and then get the calculated operator from reforigin (as well as getting the origin shifted coordinates). Then apply the operator to all phase columns of the "moving" mtz file using phasechange. It works, but as this is an unsupported programme (and can't use non-standard orthogonalizations), I'd much rather use a clipper-derived program if there is one. Thanks On Wed, Jul 31, 2013 at 12:11 PM, Kevin Cowtan <kevin.cow...@york.ac.uk>wrote: > cphasematch should do the job. I presume you did a structure factor > calculation for one or both models? > > I'm not clear what the problem is with that. Is it that it is giving you > the wrong shift? > > > On 30 July 2013 14:33, Mo Wong <mowon...@gmail.com> wrote: > >> Hi all, >> >> I am trying to get multiple molecular replacement solutions on the same >> origin. I know this has been asked before, however, in my case I want to >> stick to CCP4 programmes (I am aware PHENIX can do this). >> >> I have tried to get this to work using csymmatch which outputs the >> origin-shifted coordinates, but this gives me a solution which I can't >> apply to the reflections file (I am using phasechange to calculate the >> shifted mtz file), i.e: >> >> Change of origin: uvw = ( 0.08333, 0.5, 0.5 ) >> >> reforigin can bail due to cell dimension differences (I could recompile >> with a greater tolerance, but I'd rather not start messing with code). >> cphasematch obviously produces a sensible map with uvw=0.0,0.5,0.5, but I'd >> like to calculate the matching origin-shifted coordinates without having to >> write a fortran programme (is there a programme out there which can do >> this?). >> >> I'm guessing I can use clipper to do this in one easy step (preferably >> without having to merge the 2 mtz files in CAD), but I'd rather not have to >> re-invent the wheel if someone's done this already. >> >> Thanks for any help! >> > > > > -- > EMAIL DISCLAIMER: http://www.york.ac.uk/docs/disclaimer/email.htm >
