[gmx-users] Barostat & Density decreasing to zero

[Omer Markovitch]

Dear All,
I have simulated a protein inside a box with water and ions. I began by
minimizing my system (which has a total charge of zero), until the maximum
gradient was small enough and the potential energy become negative.
Then I heated it, slowly, to 300K at constant volume (using berendsen
thermostat and tcoupl = 1 ps).
Then I performed 500 ps NVT simulation to equilibrate the temperature, at
300K.
Then, I tried to equilibrate the pressure and run NPT simulation, using
Berendsen barostat (p=1, pcoupl=10) and Berendsen thermostat (T=300K, tcoupl
= 0.1).

My problem is, that with the barostat & thermostat both active in the same
MD run, my density drops to a very low value (at the first MD steps, the
density is 1000, dropping later to 100 kg/m^3). That is- my box dimensions
increase significantly (from around 6x6x6 to 12x12x12 nm^3. If I change the
pressure coupling constant, the speed of the expansion changes but all
simulation converged to a small density between 50-100 kg/m3.

Your help is appreciated; Probably, I am not controlling the simulation like
I should. I will appreciate any advice.
Thank you, Omer Markovitch.

I have used gromacs version 3.3.3, Here is my .MDP file for the NPT run:

integrator               = md
dt                       = 0.001
nsteps                   = 500000
comm_grps                = system
nstxout                  = 500
nstvout                  = 500
nstcheckpoint            = 1000
nstlog                   = 500
nstenergy                = 500
nstlist                  = 10
ns_type                  = simple
pbc                      = xyz
rlist                    = 1.0
coulombtype              = PME
rcoulomb-switch          = 0
rcoulomb                 = 1.0
epsilon_r                = 80
epsilon_rf               = 80
vdw-type                 = Cut-off
rvdw-switch              = 0
rvdw                     = 1.0
table-extension          = 1
fourierspacing           = 0.12
pme_order                = 4
ewald_rtol               = 1e-05
ewald_geometry           = 3d
optimize_fft             = yes
tcoupl                   = berendsen
tc-grps                  = system
tau_t                    = 0.1
ref_t                    = 300
Pcoupl                   = berendsen
Pcoupltype               = Isotropic
tau-p                    = 10
compressibility          = 4.5E-5
ref-p                    = 1
andersen_seed            = 815131
constraints              = none
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20090211/93cd93bf/attachment.html>