[ccp4bb]: ncs in refmac5

[Thomas Scott Reid <tsr@duke.edu>]

***  For details on how to be removed from this list visit the  ***
***          CCP4 home page http://www.ccp4.ac.uk         ***


Thanks for earlier advice on how to apply TLS parameters to my protein,
which has lowered both the R and Rfree by 1 and 2%, respectively.  I've
got six molecules in my asymmetric unit (one TLS per molecule), and now I
want to move on to applying both NCS restraints (which I've found, at
least in CNS, to greatly improve stereochemistry and lower the Rfree)
along with TLS restraints.  Unfortunately, I've hit a wall.

The problem is everytime I include NCS restraints, the program chokes and
gives me a 'problem with NCS' message.  However, I've found that this
problem is eliminated by getting rid of the catalytic zinc ion in my pdb
file.  Therefore, I don't think there is anything wrong with the NCS
syntax - rather, there is some other problem lurking in the shadows. Any
explanation for this phenomenon would be greatly appreciated.

Because my pdb is coming straight out of CNS, which I've been using for
some time, there must be something I'm missing in my library file.  I'm
simply letting Refmac5 read in the old CNS pdb file and then create the
appropriate .cif file.  Specifically, do I need a special library file to
describe the Zn+2 ion and its coordination sphere?  The zn ligands are
unusual - an Asp(2 bonds), Cys and His (1 bond) and a substrate ligand,
which can have either a sulfur or a nitrogen.  The zn appears in the pdb
file as follows:

ATOM   5423 ZN+2 ZN2 Z   1      26.149  91.599  68.426  1.00 45.67      Z

I have tried other notations for the Zn in the pdb file (taken from
deposited pdb files), but to no avail.

Also, does anyone know of any programs that can create
rendered/anti-aliased figures that show the Richardson's contact dots?
Xfit, although it displays them, doesn't seem to send them to Raster3D for
rendering.

All suggestions are appreciated.  Cheers.

T. Scott Reid
Duke Univ. Med. Center
tsr@duke.edu