Bond breakage during MD simulation of 2 small molecules in vacuum

GROMACS version: 2022.3
GROMACS modification: Yes/No
Here post your question

Hello everyone!

I’m trying to simulate one cyclic molecule which forms an inclusive complex with the known drug in vacuum.

The simulation runs fine when I place the drug near the cyclic molecule (sideways) but problems evolve when trying to put the drug vertically above the cyclic molecule and when making a complex in which the drug is placed inside the cyclic molecule. The same bond of the cyclic molecule becomes extremely elongated in the first step (even though starting structure looks fine) and stays that way the whole simulation.

The nvt.mdp file used for equilibration and md production:

integrator = md
dt = 0.0002
nsteps = 1500000000

nstenergy = 15000
nstlog = 15000
nstxout = 15000

constraints = hbonds

coulombtype = Cut-off
rcoulomb = 1.0
rlist = 1.0

vdwtype = Cut-off
rvdw = 1.0

tcoupl = v-rescale
tau_t = 0.1
ref_t = 300
tc_grps = System

gen_vel = yes
gen-temp = 300
gen-seed = 65823

comm-mode = angular

epsilon-r = 80

Any help will be much appreciated!