[gmx-developers] energy calculation

[Vibin R]

Dear Sir,

  This is in refrence to an earlier correspondence you had with me. I am
enclosing the mail for your refrence. 

The problem I am facing in using the wrapper is that the programme
calc_enery.c seems to be incomplete. It defines a function mcinl1100 but
does not have what actually that function is.
 Could you please throw some light on how the wrapper is to be used. It
shall be of immense help to me. I am currently working on various
optimization adaptations of the dead end elimination algorithm with a goal
towards automated protein design. I want to incorporate GROMACS to calculate
 the pairwise interaction energies for the protein side chains and also
the interaction energy of a side chain with the amino acid backbone.

Looking forward to an encouraging response,

Sincerely 
vibin


> Vibin R wrote:
> 
> > Dear Sir
> >         Can we calculate the potential energy of a structure from a
> > pdbfile with gromacs, without doing either energy mininisation or MD. If
> > it has to be done with slight adjustments in the source code, please let
> > me know, where the changes has to be made. Basically i want to get the
> > vanderwaals and coulombic interaction energies of a structure from a given
> > pdb file.
> >         Expecting your kind attention
> > Regards
> > vibin
> 
> Hi Vibin,
> 
> Together with Olivia Eriksson in Stockholm I've adapting things slightly to
> be able to use gromacs in monte carlo and sidechain energy simulations. If
> you want to try the CVS version of gromacs (that has monte carlo innerloops
> without forces) you could use the attached somewhat documented wrapper
> function to calculate the coulomb and lennard-jones energies directly. If all
> parameters are in Gromacs units you'll get the energy in gromacs units too.
> 
> However, please note that a pdb structure can have an very high potential
> energy unless you use the same forcefield as the structure was refined with.
> Normally it's a good thing (TM) to do a few steps of energy minimization to
> get around this, but your milage may vary.
> 
> Cheers,
> 
> Erik
> 
> 
> 

-- 
adios,
Manish Gupta