Hi All,
Thanks for your help with this - cphasematch is clearly what I was
looking for. For reference, here is the final script:
# compute PHI from refined structure - only necessary if these are the
# "reference" phases
sfall XYZIN 1VRM.pdb HKLIN 13185_free.mtz HKLOUT 1VRM_HKLREF.mtz << eof
mode SFCALC XYZIN HKLIN
LABIN FP=F_INFL SIGFP=SIGF_INFL
labout FC=F_CALC PHIC=PHI_CALC
eof
# cad together
cad hklin1 resolve.mtz hklin2 1VRM_HKLREF.mtz hklout combined.mtz << eof
labin file 1 ALL
labin file 2 ALL
eof
# match origins and compute mean phi difference
cphasematch -mtzin combined.mtz -mtzout dummy.mtz \
-colin-fo "[*/*/F_INFL,*/*/SIGF_INFL]" \
-colin-phifom-1 "[*/*/PHIM,*/*/FOMM]" \
-colin-fc-2 "[*/*/F_CALC,*/*/PHI_CALC]"
Cheers,
Graeme
________________________________
From: CCP4 bulletin board [mailto:[log in to unmask]] On Behalf Of
Winter, G (Graeme)
Sent: 28 August 2007 16:21
To: [log in to unmask]
Subject: [ccp4bb] Mean phase difference calculation
Hi CCP4BB,
Is there a straightforward way to compute the mean phase difference
between two phase columns in an MTZ file, allowing for the difference in
the origin? I am thinking something like:
Cad together two reflection files - renaming the PHIB column, most
likely
Compute phi offset - apply to get the origins the same
Compute the r.m.s. phase difference in degrees etc.
Also - is there a general feeling as to how useful this is for comparing
the results of phasing between different packages? I was thinking that
PHI_ref could be computed fom the final refined structure, if this is
available, as a benchmark. Alternatively is there a "correct" way to be
doing things like this?
An example script would be really great...
Thanks,
Graeme
|