*** For details on how to be removed from this list visit the ***
*** CCP4 home page http://www.ccp4.ac.uk ***
I think one of the problems was with the OXT atom that CNS includes on
the C-terminal residue. This atom was not properly refined on Refmac (I
am saying that only considering the resulting coordinates). I also did
what Partha suggested and run a few cycles of idealisation. That worked
also.
The problem now is that Refmac is giving me higher Rfree (but not
Rfactor) than CNS. I usually get better results with Refmac, that is the
first time I get worst results with it (still not running TLS).
Geommetry now is OK.
Thanks to all,
Mario Sanches
Eleanor Dodson wrote:
Is there a symmetry clash or some such serious error? When the
geometry explodes like this it is worth reading the log file carefully
to see what geometric problems are listed with the input coordinates.
Eleanor
Mario Sanches wrote:
*** For details on how to be removed from this list visit the ***
*** CCP4 home page http://www.ccp4.ac.uk ***
Dear all,
I am refining a low resolution structure (3.0A) in refmac. The
problem is that no matter which value I choose for matrix weight the
RMSD for bonds stays higher than 0.3A (yes, I didn't mistype it, it
is really 0.3A). I have tried matrix weight as low as 0.01 and also
AUTO.
Since there is two subunits in the assymetric unit I am refining
using NCS. I also tried turning it off but the problem persists. The
coordinate file I entered to refmac came from CNS were I had no
problem with geometry. Since refmac has some advantages (like TLS
refinement) I always try to change to refmac at the end of the
refinement.
Is there something I am missing? Thanks in advance for all kind of
suggestions,
Mario Sanches
