Hi,
here a brief summary regarding the use of torsion restraints with i.e.
cofactors in refmac:
The current versions of refmac have two possible pitfalls if you refine
new cofactors and have a cif-dictionary including torsion angles and planes:
Torsions are NOT automatically used in refinement. You will notice this
if you switch on monitoring with 0sig for torsions, which should show
you all torsions used in refinement. To use torsion restraints you have
to include a line like this in the input:
restr tors include resi XXX name var_1 value 180.000 sigma 5.000
period 2
This manually adds one torsion definition to the refinement. Note, that
this torsion has to be defined in the cif-dictionary (which atoms belong
to it), but the actual values are now taken from the input line! The
values in the cif are not used.
CCP4i-User: You have to use the "Developers Options" to include a
keyword file with the above lines! Otherwise your torsion will never be
used!
The second possible pitfall is, that currently the sigmas assigned for
plane definitions in the cif-dictionary are NOT used, but instead
general default values are used.
Garib was so kind as to tackle the torsion definition immediately.
There is an experimental version at
www.ysbl.york.ac.uk/refmac/data/refmac5.4_linintel.tar.gz
Now you can use the following line to include all torsions of a specific
residue:
restr tors include resi XXX
or to include only one specific definition in refinement
restr tors include resi XXX name var_1
Note that the torsion definitions have to start with "var" !
Values and sigmas will be taken from the cif-dictionary.
You can override single values explicitly by stating a full line with
new values.
The issue with the sigmas for plane definitions will be dealt with soon
by Garib.
Eckhard
--
Eckhard Hofmann <[EMAIL PROTECTED]>
Ruhr-Uni Bochum
AG Proteinkristallographie, LS Biophysik, ND04/316
44780 Bochum
Tel: +49-(0)234/32-24463, Sekr. -24461, FAX: -14762
