Re: Dihedral angle potential energy function

From: Jrme Hnin (jhenin_at_ifr88.cnrs-mrs.fr)
Date: Fri Jul 02 2010 - 07:13:10 CDT

Hi,

Before you do what you are planning, are you sure the dihedral energy
term is the information you want? The complete torsional potential
results from the "dihedral" terms, added to 1-4 LJ and Coulomb, and
some amount of additional (1-5 and further away) non-bonded
interactions.

Jerome

On 1 July 2010 12:07, Ajasja Ljubetič <ajasja.ljubetic_at_gmail.com> wrote:
> Hello,
> This might be more of a CHARMM force-field question, but since I'm more
> comfortable in the NAMD community I'll try here first :)
> In the CHARMM force-field there are dihedral angle potential functions for
> quadruples of atoms.
> So, if I'd like to plot the potential function for a rotation around a bond,
> then should I sum the potential functions for all possible quadruples around
> that bond?
> For example if I wanted to plot the potential for the rotation around the
> first bond in 2-chloropropanol then should I sum the following terms:
>
>  HO         Cl
>     \      /
> H -- C -- C -- CH3
>     /      \
>    H        H
>
> O - C - C - Cl
> O - C - C - C
> O - C - C - H
> H - C - C - Cl x 2 (each with different offset?)
> H - C - C - C  x 2
> H - C - C - H  x 2
> and in propan just
>    H        H
>     \      /
> H -- C -- C -- CH3
>     /      \
>    H        H
> H - C - C - H x 6 (each with diffrent offset?)
> H - C - C - H x 3
> Any suggestions about what would be the easiest way to obtain the potential
> for rotation around each single bond for every amino acid side-chain?
> I'm thinking of either parsing the topology file directly, getting all
> combinations of dihedrals and looking this up in the parameter file (a bit
> complicated).
> Or perhaps I could generate a peptide composed of all the difrent amino
> acids, then I have the dehidrals in the psf files, so I just look up all the
> combinations in the parameter file (still complicated but not that much).
> Perhaps I'm missing something and this is trivial to do in another way?
>
> thank you for your help & best regards,
> Ajasja

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:55:55 CST