Hi Wei,
Tough question. It depends...using GLYCAM...
1) If you have a pdb structure that contains the coordinates of the
protein and the polysaccharide, it is a relatively straight forward
exercise. Essentially you have to rename the residues to match the
Glycam residue name, which can be found in the manual (including the
N-linked amino acid to NLN). Now, if the polysaccharide is branched,
then you have to modify the input pdb to terminate each linear
carbohydrate sequence with "TER". After you have done this you will
have to create a bond between the linear sequence to create the
branched scacharide using leap and the bond command. You may have to
create a bond between the polysaccharide and the protein as well.
Please see the manual for help on this (Section 2.7.3 and Ch. 3) and
the correction for the carbohydrate example on the website (http://
amber.scripps.edu/doc8/man9_example.html).
2) Now if you are trying to create a polysaccharide chain and add to
your protein of interest, this takes a little more effort. I would
suggest that you "grow" the carbohydrate one residue at a time by
renaming the atom label within the NLN amino acid residue that will
be the C1 atom on the first attached carbohydrate(I think it is a
hydrogen) to C1 atom type. (Make sure you renumber the residue number
as well.) Execute leap and it will add the missing atoms (ie all but
C1), and thus it will put the carbohydrate in for you based on its
internal coordinate present in the prep file. Visually check to see
if there are any poor contacts between the carbohydrate and protein,
correct them by using the impose command in leap (or using the edit
command in xleap to display it graphically and manipulate it from
there). Grow the second carbohydrate by renaming the appropriate
hydrogen on the previous carbohydrate to C1 for the next carbohydrate
(again, make sure you rename the residue number). Run leap to add
the missing atoms, and visualize the output to ensure proper contacts
again. Repeat until the terminal carbohydrate is achieved. A sander
minimization on the NLN-polysaccharide will remove any erroneously
short bond distances or close contacts. This is by no means a
trivial task but it gives the most control over the formation of the
polysaccharide with respect to the protein structure.
I am sure that there is probably better methodologies out there, but
this is how my mind works out the problem. If you are having trouble
and you keep getting errors - start small by making sure you
understand how to create a polysaccharide (linear and branched) using
leap (see manual/Amber8 web page); then add your manufactured
polysaccharide to a capped NLN molecule. If you can achieve and
understand these small goals then proceed to your larger system.
Good luck,
Karl
____________________________________
Karl N. Kirschner, Ph.D.
Visiting Assistant Professor of Chemistry
Hamilton College, Clinton NY 13323
____________________________________
On Aug 17, 2005, at 2:25 PM, Wei Chen wrote:
> Dear Amber users,
> I want to attach a sugar chain to a protein, e.g. N-glycan. Does
> anybody know how I can do this?
> Thanks,
> Wei Chen
-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber.scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo.scripps.edu
Received on Wed Aug 17 2005 - 22:53:00 PDT
