[gmx-users] Regarding simulation of ions-water system

[Mark Abraham]

Hi,

On Mon, Jan 1, 2018 at 4:40 PM Dilip H N <cy16f01.dilip at nitk.edu.in> wrote:

> Hello,
> I want to simulate Ca2+, CO32- ions in the water system. So these were the
> following commands i followed:
> gmx editconf -f abc.gro -box xxx -o def.gro
> gmx insert-molecules -f def.gro -ci ghi.gro -nmol x -o jkl.gro
> gmx solvate -cp jkl.gro -cs scp216.gro -o mno.gro
> gmx pdb2gmx -f mno.gro -o mno.pdb (to generate topology)
> gmx grompp -f ions.tpr -c mno.gro -p topol.top -o ions.tpr
> gmx genion -s ions.tpr -o pqr.gro -p topol.top -pname CAL -np xx
> I hope these are the proper ways to generate topology and add ions and to
> set up the system...(Any modifications are appreciated.)
>
> And when i gave the energy minimization command (gmx grompp -f em.mdp -c
> pqr.gro -p topol.top -o em.tpr), i am getting the following error:
>

If it's an error, you'll see the word "error" and it will refuse to run
successfully. You didn't see either of these things. Some other things are
warnings or notes, when there's something you might or might not be doing
well.

Excluding 3 bonded neighbours molecule type 'Other'
> turning H bonds into constraints...
> Excluding 2 bonded neighbours molecule type 'SOL'
> turning H bonds into constraints...
> Excluding 3 bonded neighbours molecule type 'Ion2'
> turning H bonds into constraints...
> Removing all charge groups because cutoff-scheme=Verlet
>
> Atom 1 'CAL' in moleculetype 'Ion2' is not bound by a potential or
> constraint to any other atom in the same moleculetype.
>
> Atom 2 'CAL' in moleculetype 'Ion2' is not bound by a potential or
> constraint to any other atom in the same moleculetype.....
>
> how to resolve this error..?? (Is this because of the cutoff scheme..?? )
>

Not an error. Read also the message you got subsequently, but haven't
quoted here. Since you're putting multiple ions in the same moleculetype,
you would not want them to have an explicit interaction with any other ion
in the same moleculetype, right?


> Any suggestions are highly appreciated.
>

Normal practice is to put one molecule in a [moleculetype], and most people
would agree that an ion is a molecule (or at least multiple free ions are
not a molecule). All the normal force fields implement ions this way, so
you have done something atypical in setting up your topology. That's not
wrong, but you need to read the messages critically. They're just
confirming what you already know to be true - you have multiple ions in the
same [moleculetype]. The normal approach has the advantage that you can
change the number of ions in the topology by changing one number in the
[system] section.

Mark