[gmx-users] Re: gmx-users Digest, Vol 106, Issue 69

[Christoph Junghans]

> Date: Wed, 13 Feb 2013 16:07:53 +0000
> From: Steven Neumann <s.neumann08 at gmail.com>
> Subject: [gmx-users] Bond potential of atom groups
> To: Discussion list for GROMACS users <gmx-users at gromacs.org>
> Message-ID:
>         <CAKZJqQEztzuGqm0t4=f37v+W-jioiRE0pz1tvDpJ6cV7CyGLOA at mail.gmail.com>
> Content-Type: text/plain; charset=ISO-8859-1
>
> Dear Gmx Users,
>
> I want to create coarse grained model. I need bond constant and
> equilibrium distance according to the equation
>
> V(r) =K (R - Req)^2
>
> I wish to extract bonded potential between beads made out 3 atoms. I
> need Req and K. Equilibrium distance is straight forward - g_dist
> between COM or COG with index file conatining those 3 atoms. To match
> force constant K  I need to extract bonded potential and try to fit to
> my parabolic curve.
>
> My question:
>
> How can I get bonded potential between beads of 3 atoms from AA
> simulation? Do I have to specify energy groups (energgrps) before the
> run in my mdp and then use g_energy?
Have a look at the hexane example in the tutorials of the VOTCA package:
<https://code.google.com/p/votca/source/browse/?repo=csg-tutorials#hg%2Fhexane%2Fatomistic>
You will have to modify the mapping file (hexane.xml) a bit for your
case, after that you can use the csg_boltzmann program to calculate
the potential from the mapped distribution and try to fit the
resulting potential table with the above mentioned functional form.
(The VOTCA manual gives more details about these steps.)

If you have further questions contact the VOTCA mailing list:
votca at googlegroups.com

Cheers,

Christoph

>
> Thank you
>
> Steven

--
Christoph Junghans
Web: http://www.compphys.de