Properly including bromide in GROMOS96 54a7

User discussions
1

GROMACS version: 2021.3
GROMACS modification: Yes

Dear GROMACS users,

I am a beginner to GROMACS and have some previous work form a colleague available. For my purposes, I need to add bromide (Br-) to gromos 54a7 (I have a copy of the force field in my working directory).

The way my colleague has done this is by modifying only the ions.itp-file with the following:

[ moleculetype ]
; molname nrexcl
BR 1

[ atoms ]
; id at type res nr residu name at name cg nr charge mass
1 BR 1 BR BR 1 -1 79.904

I thought the form of “at type” (i.e. BR) was a bit odd compared to the other standard ions that have some kind of charge in their entry for “at type” (e.g. CL-), so when I adapted this I put BR- instead to match the syntax, thinking the omitted minus-sign was a typo. No other files of the force field have been modified by my colleague.

When I then tried to run the energy minimization step (following the Lysozyme tutorial) i got the error message “atomtype BR- not found”.

I then changed to “at type” BR instead of BR-, as my colleague used this, which resolved the error. I am however not completely sure that using BR is the correct approach. In atomtypes.atp I can see that there is bromine (non-ionic) present, which is denoted BR. Does this mean that when putting BR instead of BR- under “at type” in ions.itp, the parameters in e.g. ffnonbonded.itp for the neutral bromine are used also for the bromide? If so, is this an ok approach anyway (I can for example see that values in ffnonbonded.itp differ between entries for CL and CL-, which is my reason for doubting the approach)?

If it is not an ok practice, how do I properly add BR-? Do I have to add it also to atomtypes.atp, and to add values for c6, c12 and nonbond_params in ffnonbonded (where does one obtain such values for bromide, if so)? Are there any other files of the gromos54a7 force field that have to be modified as well?

Thank you in advance to anyone willing to sort out my confusion.

3

Hi,

The reason is that you change the name from BR → BR- and BR- is not defined as atomtype in the force field files. Atom type information are storage in ffnonbonded.itp in the [ atomtypes ] directive and [ nonbond_params ] directive (for gromos force field) and in the database atomtypes.atp.
To know more on the format see https://manual.gromacs.org/current/reference-manual/topologies/topology-file-formats.html

yes.

BR and BR- are different, it is more appropriate to describe them using different LJ parameters.

When you add a new ion, it is good to add it also in ions.itp.

\Alessandra

1 Like
4

Thank you so much for your detailed answer Alessandra!

So, if I understand you correctly I would have to do the following steps in order to properly add BR- and not use the LJ parameters for BR:

  1. Add BR- with correct values for c6 and c12 in the [atomtypes] directive of ffnonbonded.itp.
  2. Add BR- with correct values for c6 and c12 in the [nonbond_params] directive of ffnonbonded.itp
  3. Add BR- to the database atomtypes.atp
  4. Add BR- to ions.itp

Step 3 and 4 seem perfectly straightforward to me, but I am not completely sure how to do step 1 and 2.

Where would one normally find the LJ parameters for an ion, in my case bromide, if one would like to add these to ffnonbonded.itp?