Dear Camille,
For the type of molecules described .
http://en.wikipedia.org/wiki/Lipopolysaccharide
I would NOT use force field libraries (FF libs) from Glycam: This is
not because one or two sugar units, that belong to your molecule are
available in GLycam, that you have to use the Glycam FF...
Instead, I would create all required FF libs in a single & highly
homogeneous R.E.D. IV run at R.E.D. Server. Once all is parameterized,
you will have to validate your force field topology database. I think
you should split your molecule into elementary building blocks
(charged groups, fatty chains, monosaccaride units) which can be
repeated such as in a biopolymer. By providing different analogs for a
given building block, the number of possibilities is simply multiplied:
a1 b1 --> a1 b1
b2 a1 b2
b3 a1 b3
....
bn
a1 b1 --> a1 b1
a2 b2 a1 b2
b3 a1 b3
.... a2 b1
bn a2 b2
a2 b3
in this type of approach this means that you can also work on a1 or a2
only; i.e. the bn fragments can be added or not added to a1/a2; this
is quite convenient when one wants to study the impact of the
substitution/non-substitution of a particular unit by KDO for instance
you can find examples of such an approach in R.E.DD.B. The works of S.
Abel follow that approach and remain of small size; they are pretty
good examples
when on first starts:
http://q4md-forcefieldtools.org/REDDB/projects/F-92/
http://q4md-forcefieldtools.org/REDDB/projects/F-72/
then you can find more complex examples such as:
http://q4md-forcefieldtools.org/REDDB/projects/F-87/
http://q4md-forcefieldtools.org/REDDB/projects/F-85/
These projects are all related to the glycoconjugates, glycolipids...
Far larger jobs are in preparation...
regards, Francois
> Thanks for your advice! I'm kind of already doing your first suggestion in
> a rough way. My project is really in the preliminary stages and we are
> just trying to get a protocol down to be able to generate monolayer
> lipopolysaccharide surfaces on water slabs. We will need to work on the
> forcefield for the whole LPS since there are many components to it that are
> not specified in the GLYCAM parameters, but this optimization/validation
> will happen after we get the basics down. I was just curious if there was
> a KDO unit available since I thought it's inclusion in the AmberTools
> manual meant that it would be already developed, and I was afraid I was
> overlooking something since I've just started using AMBER. However if KDO
> does get released eventually, I would definitely love to know since it will
> save my lab a lot of time in the distant future.
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Fri Feb 03 2012 - 00:00:02 PST
