Dear Developers and Users
I’m going to use GROMOS in my MD simulation. In that force field, 1-4 (3rd neighbor) interactions are dealt with by particular “Sigma” and “Epsilon” potentials for interacting atoms.
I’m familiar with special_bond command, but that keyword modifies ALL interactions. On the other hand, specifying 1-4 interactions via pair_coeff modifies both intra- and inter-molecular forces.
So, Could anybody tell me “What should I do?”
Any hint is welcome.
Dear Developers and Users
I'm going to use GROMOS in my MD simulation. In that force field, 1-4 (3rd
neighbor) interactions are dealt with by particular "Sigma" and "Epsilon"
potentials for interacting atoms.
Which version of GROMOS would that be?
...and what version of LAMMPS are you using?
I'm familiar with special_bond command, but that keyword modifies ALL
interactions. On the other hand, specifying 1-4 interactions via pair_coeff
modifies both intra- and inter-molecular forces.So, Could anybody tell me "What should I do?"
do you have aromatic rings in your system. last time i checked, those
were the only excluded 1-4 interactions save for individual hydrogens,
which are uncommon, since GROMOS is a united atom force field.
Any hint is welcome.
if you want a complete and accurate representation of GROMOS, you will
need to add and modify source code to LAMMPS. i don't think LAMMPS
currently has the reaction field coulomb formulation available that
GROMOS uses, and you'd have to implement the special bond exclusions
rules. up until very recent, there was not even the gromos bond style
available in LAMMPS.
axel.
Thanks Alex.
I’m using GROMOS 53a6 and LAMMPS version 11Aug17.
I don’t have any reaction to my system.
I only want to specify 1-4 interactions explicitly. How should I do? I know special_bond modifies all interactions based on given weights and pair_coeff specifies interactions for both intra- and inter-molecular interactions.
Please notice the table 9 of the paper attached.
Many thanks in advance.
Thanks Alex.
I'm using GROMOS 53a6 and LAMMPS version 11Aug17.
I don't have any reaction to my system.
you are not paying attention, and you didn't read the paper you
attached very well.
i am not talking about reactions, i am talking about the "reaction
field method" which is an approximation to long-ranging coulomb
interactions. if you use parameters optimized for that, you cannot
simply drop in ewald summation (or PPPM).
I only want to specify 1-4 interactions explicitly. How should I do? I know
special_bond modifies all interactions based on given weights and pair_coeff
specifies interactions for both intra- and inter-molecular interactions.
as i already mentioned, you need to modify source code. the 11 Aug 17
version is missing the bond style, that GROMOS uses.
if you want to be able to provide explicit parameters for 1-4
interactions, you will have to write your own custom pair style.
CHARMM has a similar situation and has its own pair style because of
that (actually multiple, since CHARMM changed the smoothing function
for LJ interactions some time in the past).
axel.