Dear lei feng,
The cif file (GLC-b-D.cif in the Refmac dictionary) is indeed for just
one monomer. Links are specified in the list/mon_lib_list.cif file.
There you will find a description of the BETA1-4 linkage. You will see
that O4 is linked to C1 of the following residue - and that if you
examine the chem_link for BETA1-4 you will notice that O1 is deleted.
You need to make sure that the atoms in your PDB file correspond (not
all PDB models do, e.g. 3BMW does not).
Once you have consistent atoms in the PDB file, the glycosidic linkage
will be detected by Coot and just work [1]. You do not need Sketcher,
libcheck, extra cif files or extra LINK lines added to your PDB file.
Paul.
[1] I should be clear that it will "just work" with tandem unbranched
carbohydrates (as is the case in maltoheptaose).
Petr Kolenko wrote:
Hi Lei.
As I expect, you use REFMAC for structure refinement. This link should
help you to connect in principle any atoms between residues (or ligands
as well).
http://www.ccp4.ac.uk/html/refmac5/files/coordinates.html#pdb_link
You can also find many examples of different LINKs in the PDB records. I
would prefer the most similar ones. ;)
Best wishes,
Petr
Petr Kolenko
[EMAIL PROTECTED]
protein.awardspace.com
lei feng wrote:
Hello Carlos
thanks for your kindly reply. I think the cif file ( GLC_mon_lib.cif
) is only for one glucose molecule, it can not regulate the bond
between two glucose units, that is why they break apart in the real
space refinement.
in the pdb file , different glucoses are different molecules, I do
not know how to generate a cif file to give regulation on bond between
glucose units. If we can make this cif file , is it the same as
GLC_mon_lib.cif?
Thanks a lot for you help .
Yours
Lei
> Date: Wed, 5 Nov 2008 17:23:58 -0600
> From: [EMAIL PROTECTED]
> To: [EMAIL PROTECTED]
> Subject: Re: [ccp4bb] sugar ligand refinement in coot
>
> Lei,
>
> If your cif file isn't complete, you can load your pdb GLC in
Sketcher in ccp4 suite and generate a cif file. When you have your
cif, import your cif file inside coot.
>
> Carlos
>
> >>> lei feng <[EMAIL PROTECTED]> 11/05/08 4:08 PM >>>
>
> hello everyone
>
> can anyone help me with sugar ( say maltoheptaose) refinement in coot?
>
> I use the pdb file with " GLC " name , it always breaks apart into
discountinuous glucose when I try to use real space refinement , I
believe that is because there is no restraint with regarding to bond ,
whatever in cif file.
>
> I know I could give it a unique name , say G7 and substitute the "
GLC "it in the pdb file, make a new cif file and refine it , it will
not break apart again. But , when I deposit the structure in the
future , it is allowed? It seems everyone working with sugar ligands
are using " GLC" . how did they work in refinement ?
>
> Thanks for any suggestion.
>
> lei Feng
