[gmx-users] Forcefield parameters

[Mark Abraham]

----- Original Message -----
From: Sai Pooja <saipooja at gmail.com>
Date: Wednesday, October 27, 2010 8:52
Subject: Re: [gmx-users] Forcefield parameters
To: Discussion list for GROMACS users <gmx-users at gromacs.org>

> On Tue, Oct 26, 2010 at 4:04 PM, Justin A. Lemkul <jalemkul at vt.edu> wrote:
 > 
> 
> Sai Pooja wrote:  
 > 
> 
> On Tue, Oct 26, 2010 at 3:46 PM, Justin A. Lemkul <jalemkul at vt.edu <mailto:jalemkul at vt.edu>> wrote:
  > 
> 
> 
>    Sai Pooja wrote:
> 
>        Hi,
>         I want to change the non-bonded parameters to modify the
>        interaction between water molecules and protein molecules.
>         I am using CHARMM forcefield with Tip3p water.
>           The ffnonbonded.itp file of the forcefield has non-bonded
>        parameters for tip3p water. Can I achieve the above by changing
>        these parameters?
>         
> 
>    That depends on your definition of "modify," but yes, in a way, you
>      can make changes here. 
>   1) Modify - Multiply sigma and epsilon by a constant
> 
>        If yes, will this also change the non-bonded parameters for
>        water - water interaction?
> 
> 2) Is there a way to add a new ifdef perhaps such that a modified sigma and epsilon can be used for water-protein interactions and the unmodified parameters can be used for water-water interactions?
>    
> 
> Nonbonded interactions are calculated during the simulation by applying the combination rules defined by the force field.  There is no simple way to do this with an ifdef, since that is just in the topology.  You can't conditionally apply nonbonded parameters.  That just sounds like a recipe for breaking a force field.

Not quite right. Parameters for VDW are calculated from the combination rules from the atom-specific values given in [atomtypes] only as a last resort. [nonbond_params] are used in preference to such.

So modified protein-water VDW interactions can be introduced by defining all relevant "protein atom"-"TIP3P oxygen" [nonbond_params] terms. It may be simpler to modify the [atomtypes] to generate the "modified" VDW from the combination rule, and introduce the "normal" TIP3P oxygen-oxygen interaction via [nonbond_params].

Mark

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20101027/7bb5773c/attachment.html>