AMBER Archive (2009)

Subject: RE: [AMBER] question

From: Ross Walker (ross_at_rosswalker.co.uk)
Date: Wed Jun 10 2009 - 22:19:01 CDT


Dear Lev,

> I need to run a specific MD simulation with the following features:
>
> (1) periodic boundary conditions (2D or 3D); essentially, I have a
> periodic monolayer of Guanine molecules

You can do 3D periodic boundaries, with PME, with AMBER, You cannot do 2D so
you would want to either stack multiple layers or solvate the system top and
bottom.
 
> (2) there are 8 molecules in the unit cell forming two tetrads, each
> tetrad consists of 4 molecules forming a
> square so that you can imagine that the unit cell consists of two
> squares with 4 molecules each;
> we would like to run an MD simulation
> with the angle between the two squares fixed. The angle cannot be
> defined by just 3 atoms, more atoms should be
> involved (e.g. 4 atoms from each tetrad, 8 altogether), however, many
> definitions can be given. Is it possible?

I believe you can do this either using some of the NCSU modules or some of
the generalized plane-plane restraints. I have not tried this myself but
Mathew Seetin or Volodymyr Babin can probably comment some more.

Note for a periodic boundary representation you may need to include multiple
unit cells since the minimum box size must be at least half the cut off
(which has a minimum of 8) to avoid issues with the direct space sum
including the images.

> (4) can one calculate the vibrations for the given geometry? Most
> likely
> - yes.

This depends on what you mean here. You can do a normal mode analysis but
that is based on an optimized geometry. You can also save geometry vs time
so you could calculation correlations etc.
 
> (5) are H-bonding interactions between Guanine molecules described
> reasonably well? Which force
> field (from those implemented) should be best for that system?

I think most people would suggest using FF99SB for protein systems, this
includes the FF99 parameters for DNA. You might also want to try the BSC0
parameters. As for H bonding, it should work reasonably well but for a
definitive comment I would suggest checking the literature, beginning with
the 99 and BSC0 force field papers and see what is discussed here.

Good luck,
Ross

/\
\/
|\oss Walker

| Assistant Research Professor |
| San Diego Supercomputer Center |
| Tel: +1 858 822 0854 | EMail:- ross_at_rosswalker.co.uk |
| http://www.rosswalker.co.uk | PGP Key available on request |

Note: Electronic Mail is not secure, has no guarantee of delivery, may not
be read every day, and should not be used for urgent or sensitive issues.

_______________________________________________
AMBER mailing list
AMBER_at_ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber