This may also be a job for 'convert2mtz', which I think will pick up the
complex columns automatically. Only available from the command line though.
Kevin
Eleanor Dodson wrote:
I am no expert on CNS files, but do the entries for F_BULK and F_MODEL
correspond to an amplitude and a phase - it certainly looks like that..
If so you can read both in:
I havent checked the format but your script would look like this
title [No title given]
format
'(6X,3F5.0,6X,F10.0,6X,F10.0,6X,F10.0,/,25X,F10.0,6X,F10.0,7X,2F10.0,/,28X,F10.0,10X,F10.0,10X,F10.0,/,28X,F10.0,11X,F10.0,/,27X,2F10.0)'
skipline 5
labout H K L IP SIGIP FP SIGFP FreeRflag Fbulk PHIbulk HLA HLB HLC HLD
FOM Fcalc PHIcalc
ctypout H H H J Q F Q X F P A A A A W F P
PNAME tony
DNAME z
XNAME x
end
Miriam Sharpe wrote:
Hi everyone,
I've been looking at ways to convert an old hkl file from a CNS
refinement (see details below) into an mtz file.
F2mtz in ccp4 will do this once I have told it the fortran format of
the hkl file (below), and specified the data type and label for the
input fields.
However, I'm not sure how to define the F-MODEL and FBULK columns in
terms of mtz column types. Both are two-part columns containing
complex Fcalc with both amplitude and phase information.
Any suggestions are most welcome!
Thanks,
Miriam Sharpe
Dr Miriam L Sharpe
Postdoctoral Fellow
Department of Biochemistry University of Otago
PO Box 56
710 Cumberland St
Dunedin 9054 , New Zealand
_______________________________________________________________
the list of columns in the hkl file:
IOBS=experimental intensity from .sca file
SIGI=sigma of IOBS
FOBS=square root of IOBS
SIGMA=sigma of FOBS
TEST=Rfree flag
FBULK=bulk solvent part of Fcalc
PA_MODEL = HL coeff.
PB_MODEL= HL coeff.
PC_MODEL = HL coeff.
PD_MODEL = HL coeff.
FOM_MODEL =figure of merit
F_MODEL = complex Fcalc, includes both amplitude and phase
and the first two reflection entries:
INDE 38 21 4 IOBS= 771.500 SIGI= 1140.300 FOBS= 27.780
SIGMA= 27.780 TEST= 0 FBULK=
0.333 55.023
PA_MODEL= 5.685 PB_MODEL= 5.558
PC_MODEL= 0.000
PD_MODEL= 0.000 FOM_MODEL= 0.935
F_MODEL= 135.595 44.354
INDE 38 21 3 IOBS= 18.500 SIGI= 861.100 FOBS= 4.300
SIGMA= 4.300 TEST= 0 FBULK=
0.542 82.489
PA_MODEL= 1.334 PB_MODEL= 0.159
PC_MODEL= 0.000
PD_MODEL= 0.000 FOM_MODEL= 0.555
F_MODEL= 147.826 6.802
Corresponding Fortran format (I think):
(6X,3F5.0,6X,F10.0,6X,F10.0,6X,F10.0,/,25X,F10.0,6X,F10.0,7X,2F10.0,/,28X,F10.0,10X,F10.0,10X,F10.0,/,28X,F10.0,11X,F10.0,/,27X,2F10.0)
MTZ Column Types:
H
index h,k,l
F
structure amplitude, F
D
anomalous difference
Q
standard deviation of anything: J,F,D or other
P
phase angle in degrees
W
weight (of some sort)
A
phase probability coefficients (Hendrickson/Lattmann)
B
BATCH number
S
1/d**2 = 4 (sin theta/lambda)**2
J
intensity
Y
M/ISYM, packed partial/reject flag and symmetry number
I
any other integer
R
any other real
U
partial FC (see CAD)
V
partial PHIC (see CAD)
