[gmx-users] GROMOS-96 proper dihedrals

[Gerrit Groenhof]

Hi Dallas,


> Another proper dihedral issue .... I get the feeling this has been brought
> up previously, but I can't happen to find it in the archives at the
> moment.  If you have more than one multiplicity and force constant for a
> proper dihedral, can you simply define two for the same proper dihedral?
> For example multiplicity of 1 and 2 with different force constants for each
> multiplicity.


You can do that. Just add an extra line with the same 4 atoms that
define the first dihedral and use another set of dihedral forcefield
parameters. Look for instance at the pyrophosphate (or something) part
of the NADH entry in the ffG43**.rtp. 

[ dihedrals ]
;  ai    aj    ak    al   gromos type
 AO5*  AC5*  AC4*  AC3*     gd_17   
 AO5*  AC5*  AC4*  AC3*     gd_7    

However, there was an issue with pdb2gmx that should have been fixed
by now. The faulty version was not reading in two times the same
dihedral from the rtp file. So if you're using pdb2gmx and you have
defined your residue with two different dihedrals over the same
torsion in the rtp file, check in the topology if there are indeed two
propers!

Good luck,

Gerrit