[gmx-users] Questions concerning Gromacs

[Mark Abraham]

On 5/01/2011 6:14 PM, Thomas Koller wrote:
> Hello,
>
> I have some questions concerning the implementation in Gromacs:
>
> 1) I want to model a rigid anion of an IL (fixed bonds and angles). The cation is flexible. How can I do that without using constraint types? With LINCS or SHAKE, I can only fix, for instance, bonds with H atoms or all atoms.

Angles can be constrained. See manual 7.3

>   Can I use high values for the force constants?

You can use anything you can validate...

> 2) When implementing improper dihedrals, I write the sequences of the four atoms in the [dihedraltypes] and then fix them in the [dihedral] directive. But I always get the error that the impropers are not defined in the default. What is my problem?

There are no parameters found. Unfortunately we can't tell why that is so.

> I use function type 2. The sequence of impropers is: center atom - neighbor atom - neighbor atom- neighbor atom,
> as I saw in the manual.
>
> 3) For function type 1 of the dihedrals, I implement the equilibrium angle, the force constant and the multiplicity (one value for all kinds). What should I do if I have, for instance, three equilibrium angles, different for the multiplicities (n=1,2,3). How can I implement that in the top file?

If you have functions that are meant to add together on the same 
dihedral, then simply list them and they will be added.

You've got a much better chance of useful feedback if you can provide 
short, simple examples of what you are doing, and what feedback you 
don't like.

Mark