[gmx-users] parameters for CH1 -- S -- CH2 ?

[Mark Abraham]

On 20/08/2011 1:57 AM, Yun Shi wrote:
> Hi all,
>
> I have a small molecule with a part like:
>
> CH1 --- CH1 --- S --- CH2 --- CH1
>              |
>              CH1
>
> I first tried PRODRG, and it turned out PRODRG assigned bond, angle, 
> and dihedral parameters according what are already present in the 
> gromos53a6.ff/ffbonded.itp file. Since this file does not contain 
> information regarding this CH1 --- S --- CH2 pattern, PRODRG just 
> assigned bond (gb_31),angle (ga_4) as if my molecule has CH3 --- S --- 
> CH2 pattern, and dihedral parameters for CH1 -- CH1 -- S -- CH2 as if 
> it was CH1 -- CH2 -- S -- CH2.
>
> So I guess I have to do this manually. But as no corresponding 
> parameters defined in ffbonded.itp file, should I derive these 
> parameters via chemical intuition and try validate them in some way?
>
> Or is there an easier alternative way to do this?

Parameterizing new species is not easy. See 
http://www.gromacs.org/Documentation/How-tos/Parameterization. Be sure 
to check the literature for existing approaches - and particularly how 
the thioether parameters that do exist were derived.

Mark