[gmx-users] topology peptide plane

[Mark Abraham]

abelius wrote:
> Dear All,
> 
> I'm trying to write some scripts for converting topology files.
> But I need some more info in the used formats.
> 
> My question is how can I define a peptide plane in a topology include file
> lets say my system is:
>  
>  O(3)       C(5)
>   \\           /
>    C(2) - N(4)
>  /             \
> C(1)          H(6)

I think the concept you are looking for is that of an "improper 
dihedral". In the above example, an additional dihedral "angle" between 
atoms 6-2-4-3 that was harmonic with a minimum at 180 degrees will 
(help) keep the group planar. In practice you'd use a few of these. Look 
in the gromacs manual for more description, and look at the impropers in 
the other peptides in the force field files. It would also be a good 
idea to read the papers that describe the development of the force 
fields you're converting from and to, to make sure you are staying 
consistent with them.

Mark