*** For details on how to be removed from this list visit the ***
*** CCP4 home page http://www.ccp4.ac.uk ***
Dear All, and especially Lena and Christian,
I recently had a couple of structures where the active site contained
a mixture of either product and intermediate or substrate and
product. Like Christian, when I defined the compounds as separate
ligands the unchanging portion of the ligands refined badly in refmac
- this was solved by defining the ligand mixture as a fixed portion
with an occupancy of 1 and linking this to two different ligand
portions with partial occupancy (totalling 1). It took some tweaking
to figure out exactly where the link should be (and consultation with
Garib about how to get correctly define the compounds and the links),
but managed to get good refinement from refmac. One thing to look
out for with this approach is that in the dictionary the ligand
portion is completely defined (eg a hexose and a nucleotide) and the
link is defined, but in addition you need to modify the compounds so
that the link can be formed (eg removing a H).
in the .cif file you need: a list of monomers, a list of links, list
of modifications, description of monomers, description of
modifications, description of links
in the pdb file after the last REMARK line, before CISPEP and
CRYST1, have lines describing the links, here is an example:
LINK O3A GDP A 400 PB GUL A 401 GDPGUL
this links the O3A of GDP to PB of a hexose mono phosphate, this link
is defined in the .cif file as GDPGUL.
One other thing that I needed to play with was the position of the
link - eventually found that the guanosine, ribose and one phosphate
were in fairly constant positions, the 2nd phosphate moved
significantly depending upon which sugar moiety was in the active
site. Additionally, I did need to check the cif file bond lengths,
angles, torsion and planarity values to get sensible sugars out of refmac.
This strategy was worth the effort - ended up with refined ligands
that were a good match for the initial unrefined density.
Good luck,
louise
********************************************************
Louise Major
Centre for Biomolecular Sciences
North Haugh
The University
St Andrews
Fife KY16 9ST
Scotland
ph +44 1334 467282
fax +44 1334 462595
