GROMACS version: 2020.3
GROMACS modification: No
Here post your question
Greetings! I’m a neophyte trying to port a new force field into gromacs. I have a quick question about atom names in the [ atomtypes ] for a ffnonbonded.itp in a force field.
In the user manual, the atom name at the very beginning of an atom definition contains a short string for identifying the atom before the interger number used to designate the element number. I’ve noticed that certain general force fields present with gromacs actually include two names here, particularly OPLS-aa. This field has an atom name, then a bond type name, and then the element number. As far as I can tell, the second name ties the entry in the ffnonbonded.itp to sister entries in the ffbonded.itp file and seems to act as a sort of wild card to allow a single bonded entry cooperate with a number of different nonbonded types.
I can’t find anything in the gromacs manual about how to instruct gromacs as to whether or not the second name should be included, so I must assume that either gromacs looks for multiple names automatically, or that these names must be defined somewhere that I haven’t found. What are the limits of how this can be used? Can I add ten names before the integer? Does it only look for the first name under certain circumstances and the second name in others, like using the first name to identify VdW interactions and the second name for occurrences in the bonded interactions? If I add a [ nonbond_params ] directive for including special non-diagonal interactions, bypassing combination rules, must it refer to the first name and not the second or does it not care?
If anyone can give me some guidance on this, I would be appreciative. Thank you very much!