Dear Jana,
> I wish to perform MD simulations on a triple molecular system using
> Amber version 10:
> Protein-double stranded DNA-a deoxynucleotide triphosphate.
> Double stranded DNA contains lesions.
>
> I would greatly appreciate if:
> - you could share the force fields parameters files for:
> 1. Deoxynucleotide triphosphate, preferable, dATP
> 2. cyclobutane thymine dimer
> 3. N2-furfuryl-dGTP
I do not think you will find all what you need in the AMBER parameter
database...
> OR
>
> - you could provide advice where I can obtain the files in such
> suitable format which would allow me to load these files directly in
> tleap.
You could use mol2, OFF or prep file format to load FF libraries
(these days the tendency is to use the mol2 file format) and frcmod
file to load FF parameter files in LEaP (see the LEaP documentation).
For the structures which are not in the AMBER parameter database - or
in R.E.DD.B., you should start by generating a corresponding FF
library. Here, you could follow two different approaches: (1) you
study the molecule constituted by a single "block" (2) or you split
your molecule into "fragments".
* When you follow the approach (1) you generally face to three major problems:
- for big molecules with a complex potential energy surface the
conformation used in the charge derivation is often not well defined
(& generally you want to control this conformation)
- if your molecule is multiply-charged the minimum generated might not
be correct or difficult to obtain.
- if you decide to follow the approach (1) & if your structure is part
of a biopolymer for instance, you will have to generate a fragment for
your structure anyway; to be able to physically connect it to this
biopolymer.
* Then, the second idea is to split your molecule into fragments -
approach (2). This is the basis of the Amber and GLYCAM force field
topology databases where you build a polymer based on multiple
fragments which are combined in LEaP. However, obtaining molecular
fragments is often seen as more complex.
- If you want to follow the approach (1) you can use Antechamber or R.E.D.
- If you want to follow the approach (2) I suggest you to use R.E.D.
and even R.E.D. Server. You will find many examples of such fragment
based approach in R.E.DD.B. (more or less complex), and tutorials are
also available.
Please, see:
http://q4md-forcefieldtools.org/RED/
http://q4md-forcefieldtools.org/REDS/
http://q4md-forcefieldtools.org/REDDB/
http://q4md-forcefieldtools.org/Tutorial/
I tried to make my answer as short as possible, but obviously this is
just a starting point, and more could be added. Do not hesitate to ask
more questions about the parametrization of your structures if needed.
regards, Francois
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Sat Jan 30 2010 - 03:00:03 PST
