[gmx-users] Modifying gromacs

[Sai Pooja]

Hi,

I want to modify the contribution to Interaction energy of different groups
- say I have groups A and B and I want the energy to be scaled as E_AA +
0.1E_AB + 0.5*E_BB. Interaction parameters of each of these groups are set
by a forcefield

Now after multiple correspondence on the gromacs list I have concluded that
there are 3 ways of doing this:

1. Using tables - for this I would have to list non-bonded parameters for
all atoms such that the combination rule and the table-potential is used.
For the table for BB interaction, scale the Coloumb and VdW interactions in
the tables by a factor of 0.5 and so on...
However, since tables would have to be supplied for pairs too (tablep), it
may not be accurate to supply 6-12 tables with coulomb potential for these
pairs. I am using CHARMM and the 1998 paper on CHARMM says that in some
specific cases the 1-4 interactions many be scaled which makes me doubt this
approach.

2. Forcefield parameters - By defining scaled [nonbonded_params] for all
relevant atoms. This will change the VdW interactions, but not sure about
the Coulomb interactions.

3. Modifying gromacs - by passing a parameter lambda to gromacs which scales
the force/potential by a factor lambda when gromacs calculates
force/potential.
For implementing option 3, which programs in the gromacs package would be
the bes tstarting points for editing the energy contributions of different
groups/atoms?

As a more general question, how does one run a generalized Hamiltonian REM
on gromacs?

Pooja



-- 
Quaerendo Invenietis-Seek and you shall discover.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20101028/b173ccc9/attachment.html>