[lammps-users] protein as rigid bodies

Hi, i’am new in lammps. i try to simulate 2 proteins as rigid bodies(each of them has a layer of water 3A, total numbers of atoms almost 8000) in order to approach each other. i used fix_rigid command for proteins but simulation runs too slowly(like an all atom figuration)almost 10 steps per 25sec. could you send me an appropriate input script about that. thanks in advance.

Lots of things can make a simulation run slowly - e.g. long cutoff,
hi-precision long-range Coulombics, etc. Does it run any differently
when you don't treat the proteins as rigid bodies? If not, then
fix rigid has nothing to do with the speed.

Steve

Hi, i'am new in lammps. i try to simulate 2 proteins as rigid bodies(each
of them has a layer of water 3A, total numbers of atoms almost 8000) in
order to approach each other. i used fix_rigid command for proteins but
simulation runs too slowly(like an all atom figuration)almost 10 steps per
25sec. could you send me an appropriate input script about that. thanks in
advance.

fix rigid will not change how you compute the forces, so the computational
effort will be the same as in an all-atom simulation.

what you seem to be looking for is "coarse graining".
there are a number of approaches to that (including one
that is being actively developed in our group at temple
and supported in LAMMPS. hint, hint ;)).

cheers,
   axel.