[gmx-users] Box size problem

[Davit Hakobyan]

Dear All,

I try to run a simulation with protein-membrane system in charmm36 ff 
with Gromacs5.0.4 and the simulation runs without complain.The problem 
is in the unit cell size which seems to grow during a short 500 ps 
simulation from ~15 to ~28 nm in Z direction although the system atoms 
remain confined in the~15 nm in Z direction.


The GRO files before and after the 500 ps simulation can be downloaded 
from the following link: 
https://uni-muenster.sciebo.de/index.php/s/GBHjED2t1KzZZ2o

And these are the MDP parameters used:

MDP Parameters
-------------------------------------------------
integrator = md
dt                      = 0.001
nsteps                  = 500000

nstxout                 = 500000
nstvout                 = 500000
nstlog                  = 100000
nstenergy               = 500000
nstxtcout               = 500000
xtc_grps                =
nstcalcenergy           = 500

energygrps              = ANA2 CHL1 POPC DOPC POPI POPS CAL

cutoff-scheme           = Verlet
nstlist                 = 20
rlist                   = 1.2
coulombtype             = pme
rcoulomb                = 1.2
vdwtype                 = Cut-off
vdw-modifier            = Force-switch
rvdw_switch             = 1.0
rvdw                    = 1.2
;
tcoupl                  = Nose-Hoover
tc_grps                 = PROT   MEMB   SOL_ION
tau_t                   = 1.0    1.0    1.0
;ref_t                   = 303.15 303.15 303.15
ref_t                   = 200. 200. 200.
;
pcoupl                  = Parrinello-Rahman
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               = PROT   MEMB   SOL_ION
;
refcoord_scaling        = com
-------------------------------------------------

No evident problem was displayed at TPRpreparation or MD run.


Could some combination of parameters cause such a problem?
With Gromacs 5.0.4 I have successfully simulated similar 
protein-membrane systemsin the past with the same charmm36 force field, 
but at that time different parameters I remember were suggested in 
Gromacs, particularly, the following ones:

MDP Parameters (OLD)
-------------------------------------------------
integrator          = md
dt                  = 0.002
nsteps              = 250000000

nstxout             = 500000
nstvout             = 500000
nstlog              = 100000
nstenergy           = 500000
nstxtcout           = 500000
xtc_grps            =

energygrps          = Protein CHL1 POPC DOPC POPS Water_and_ions
;energygrp_table    =
nstcalcenergy       = 500
nstlist             = 10
nstcomm             = 50
comm_mode           = Linear
comm-grps           =
ns_type             = grid
rlist               = 1.0
rlistlong           = 1.4
rvdw_switch         = 0.8
vdwtype             = Switch
coulombtype         = pme
rcoulomb            = 1.0
rcoulomb_switch     = 0.0
rvdw                = 1.2
fourierspacing      = 0.15
pme_order           = 6
;ewald_rtol         = 1e-6
tcoupl              = V-rescale ;nose-hoover
nhchainlength       = 1
tc-grps             = Protein CHL1 POPC DOPC POPS Water_and_ions
tau_t               = 0.1 0.1 0.1 0.1 0.1 0.1
ref_t               = 293 293 293 293 293 293
Pcoupl              = parrinello-rahman ;berendsen ;parrinello-rahman
Pcoupltype          = semiisotropic
tau_p               = 5.0
compressibility     = 4.5e-5 4.5e-5
ref_p               = 1.0 1.0
pbc                 = xyz
gen_vel             = no ;yes
gen_temp            = 293
optimize_fft        = yes
constraints         = hbonds
continuation        = no
constraint_algorithm = Lincs
lincs-order         = 4 ; 8 is needed for BD with large time-steps.
lincs-iter          = 1 ; 1 is fine for normal simulations, but use 2 to 
conserve energy in NVE runs.
-------------------------------------------------

In the new parameters I also tried to change the tcoupl from Nose-Hoover 
to V-rescale, but the outcome is the same, the box size becomes large.

ProbablyI am missing something trivial.
Does anybody have a clue what might be wrong?

Thanks very much for any help.
Davit