[gmx-users] parameters for bond types for GROMOS force field.

[Justin A. Lemkul]

On 6/10/12 3:06 AM, James Starlight wrote:
> Justin
>
> I have one extra question about parametrisation of the bond type
>
> Initially I had -c-c- bond so both atoms were in the sp3 hybridization. I've
> changed bond type between that atoms to the gb_16 ( from phe ring) to define
> them as the >c=c< but during the simulation I've noticed that the bond between
> that atoms is not planar ( sp2 form) and still is in the sp3. What addition
> changees should I do in my topology to change the planarity of new bond ?
>
>

Defining appropriate atom types will be necessary here, as well.  For a single 
bond, the atom types would be -CH2-CH2- while a double bond will necessarily be 
-CH=CH-

Also note that with a united atom representation, it's very hard to make such 
assertions about hybridization state.  There are no hydrogens, so the geometry 
of these central carbons is not easily defined.

> Should I make some changes ia angles term as well?
>
> E.g I've found
>
>    612   613   614     2    ga_13  ;
>
> where 612 and 613 are the atoms wich I would like to present as the rigid. I've
> change to the ga_27 ( 120) but during simulation the bond was still in sp3. What
> I've done wrong ?
>

As there are no parameters for linear C=C bonds in the Gromos force fields, you 
may have to do some parameterization yourself, particularly for appropriate 
dihedrals.  One option is to look at how these bonds are treated in lipid force 
fields like the Berger parameter set, distributed by Peter Tieleman's group. 
There are double bonds defined for lipids like POPC.

-Justin

-- 
========================================

Justin A. Lemkul, Ph.D.
Research Scientist
Department of Biochemistry
Virginia Tech
Blacksburg, VA
jalemkul[at]vt.edu | (540) 231-9080
http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin

========================================