[gmx-users] On CHARMM angle bending potential

[David van der Spoel]

On 04/07/16 12:11, Mr Bernard Ramos wrote:
> Hi all!
>
> The potential for the angle-vibration potential in CHARMM is V = k(a - a0)^2. Its corresponding force field is proportional to 1/sin(a) which blows up when a approaches \pi or when the molecule is in linear arrangement. I am wondering how Gromacs calculates the angle-bending potential of a linear molecule such as carbon dioxide when it the force vector is singular at a equal to \pi?
>
> Thank you.
>
> Bernard
>
It depends on the force field. Gaff sets the reference angle to 178 or 
so. An alternative is to define the C as a virtual site, which is 
probably easiest to implement in most force fields, but only practical 
if the angle fluctuations are irrelevant for the problem at hand. My 
group are working on more systematic way of treating angle potentials, 
and one such linear angle is implemented in gromacs already, however it 
has not been parameterized rigorously and published.

-- 
David van der Spoel, Ph.D., Professor of Biology
Dept. of Cell & Molec. Biol., Uppsala University.
Box 596, 75124 Uppsala, Sweden. Phone:	+46184714205.
spoel at xray.bmc.uu.se    http://folding.bmc.uu.se