*** For details on how to be removed from this list visit the ***
*** CCP4 home page http://www.ccp4.ac.uk ***
SHELX was never intended for refinements at lower resolution than 2A, it
tends to 'over-refine' and give a low R-factor at the cost of poorer
geometry and B-values unless used very carefully with tight restraints
(i.e. small esds). It does not provide torsion angle restraints, this
has the advantage that you can use the Ramachandran plot etc. for
diagnostic purposes (as you are correctly doing). However if you have
NCS you should try using the NCSY non-crystalographi symmetry
restraints, which effectively restrains equivalent torsion angles in
different molecules to be equal. Including the hydrogen atoms on carbon
and nitrogen in combination with the antibumping restraints will also
help to avoid bad torsion angles, so with care it may be possible to get
a reasonable refinement with SHELX (do not include the hydrogens of OH
groups in such cases, they can cause problems and do nothing to help you).
In fact your R and free-R are very respectable for such a case, though
care is needed in the selection of the free-R reflections for twins.
As far as I know, Refmac can not yet refine twins, but I understand that
Garib is working on it. CNS can, but you will need advice from a CNS
expert. In fact at this resolution CNS is a good choice anyway and would
allow NCS constraints and torsion angle restraints.
George
Ethayathulla Abdulsamath wrote:
hi all
I am refining twinned data 2.5A resolution using SHELX. My problem it's
geometry get bad I mean dihedral angle (ramachandran plot ). It keeps on
decreasing when i try to refine further to imporve my phases. Is there
anyway to define restrain for this dihedral angles.
I am notable to refine same structure using ccp4 and cns using same
twinning operator and twinning fraction. Any comment on this .
The R-factor is 21 Rfree is 25 in shelx whenever i tried to refine using
same twinning fraction which is already refined using shelx in both ccp4
and cns it gives around R & Rfree of 33 & 34. That means it is not
taking account of twinning fraction I think so . but iam following same
procedure whichever given in CNS & CCP4 manual. But it is not working
for me.
But shelx works well but most allowed region decrease refinement by
refinement.
plz give suggestions on this !
Thanking you
Ethayathulla
###################################################
A.S.Ethayathulla,Ph.D.
Department of Biophysics
All India Institute of Medical Sciences
Ansari Nagar
New Delhi-110029
India.
###################################################
<http://adworks.rediff.com/cgi-bin/AdWorks/sigclick.cgi/www.rediff.com/signature-home.htm/[EMAIL PROTECTED]>
--
Prof. George M. Sheldrick FRS
Dept. Structural Chemistry,
University of Goettingen,
Tammannstr. 4,
D37077 Goettingen, Germany
Tel. +49-551-39-3021 or -3068
Fax. +49-551-39-2582
