Comparing Simulations with Different Thermo/Barostats

GROMACS version: 2023.5
GROMACS modification: No
Hi! I’ve been simulating different large protein-membrane systems (> 500k atoms) for a while now to investigate the oligomerisation and membrane interaction of my protein of study. I’ve always prepared the systems with CharmmGUI, but I just noticed that my most recent simulations run with different thermo/barostats to the previous simulations. They switched from Nose-Hoover/Parrinello-Rahman to v-rescale/c-rescale. The rest of the parameters in the mdp file are the same. Here is my latest mdp version:

integrator              = md
dt                      = 0.002
nsteps                  = 500000
nstxout-compressed      = 50000
nstxout                 = 0
nstvout                 = 0
nstfout                 = 0
nstcalcenergy           = 100
nstenergy               = 1000
nstlog                  = 1000
;
cutoff-scheme           = Verlet
nstlist                 = 20
rlist                   = 1.2
vdwtype                 = Cut-off
vdw-modifier            = Force-switch
rvdw_switch             = 1.0
rvdw                    = 1.2
coulombtype             = PME
rcoulomb                = 1.2
;
tcoupl                  = v-rescale
tc_grps                 = SOLU MEMB SOLV
tau_t                   = 1.0 1.0 1.0
ref_t                   = 303.15 303.15 303.15
;
pcoupl                  = C-rescale
pcoupltype              = semiisotropic
tau_p                   = 5.0
compressibility         = 4.5e-5  4.5e-5
ref_p                   = 1.0     1.0
;
constraints             = h-bonds
constraint_algorithm    = LINCS
continuation            = yes
;
nstcomm                 = 100
comm_mode               = linear
comm_grps               = SOLU_MEMB SOLV

While I’ve seen on the forum that v-rescale and c-rescale are highly recommended now, I worry that my simulations may not be comparable between them. For instance, if we publish a paper including results from two or more protein-membrane systems (such as the wild-type protein and a mutated version, or different order oligomers, or membrane placements) that have been simulated with different thermo/barostats.

Should I worry about this or is this not a fatal mistake?

Thank you in advance.

All thermostats and barostats, assuming the algorithms are correct and parameters reasonable, should produce the same ensemble. So the choice should not affect the results. With that said, second order algorithms, such as Nose-Hoover and Parrinello-Rahman, can cause instabilities, so that is why we recommend first order coupling.

Well that’s very good news then. Thank you for your answer. Have a good day! :)