Somewhat unrelated to your question, but relevant to what you are doing:
I was once told by an expert that NDG is not known to occur much in
glycoproteins; it is usually an NAG that was built using the wrong
stereoisomer by a careless crystallographer. I have actually looked at a
few higher-resolution pdbs that included NDG monomers, and whenever the
density was clear it was also clear that the NAG was incorrectly built
as NDG. Similarly, you should have alpha-linked fucose, which is FUC,
not beta-linker, which is called FUL (Also, BMA [beta] and MAN [alpha]
are commonly mixed up). I do not know the specifics of this pdb you are
mentioning, but you may want to fix those several issues. Also, see
previous posts on this matter.
I would strongly recommend against using somebody else's sugars from a
pdb. Always build your own. Refmac will recognize your sugars and
linkages if you use the correct monomers and linkage types, because they
are all in the standard monomer library that ships with ccp4 (and phenix
and cns). I believe CNS 1.3 actually deduces (and may even correct) the
linkages from your model pdb, which I hope refmac and phenix.refine will
implement eventually.
Good luck,
Engin
On 1/10/11 3:02 PM, Hailiang Zhang wrote:
Hi,
I am running refmac on gp120(PDB 3FUS), and wondering whether there are
any dictionary files (.cif) that have already been built for the
polysaccharide (containing FUL BMA MAN NAG NDG, with NAG linked to ASN).
Thanks in advance for any help!
Best Regards, Hailiang
--
Engin Özkan
Post-doctoral Scholar
Howard Hughes Medical Institute
Dept of Molecular and Cellular Physiology
279 Campus Drive, Beckman Center B173
Stanford School of Medicine
Stanford, CA 94305
ph: (650)-498-7111
