[gmx-users] Pull code problem: Large fluctuations in pull force in Gromacs 4.5.1 compared to 4.0.5

[Emma Eriksson]

Dear all,
I have a problem using the pull code in Gromacs 4.5.1. I am running simulations with a small molecule at constrained distances from the bilayer center in a membrane (z-direction) in order to calculate permeability data. I use the force acting on the molecule (pullf.xvg) at different distances to calculate free energy profiles and diffusion coefficients.
The problem I am facing is that the pull forces are fluctuating significantly more in version 4.5.1 compared to in 4.0.5. In 4.0.5 the forces normally fluctuate in the range of at most ±2000 kJ/mol/nm whereas in 4.5.1 they fluctuate in the range of at most ±50000 kJ/mol/nm. The average force is however similar in both versions, which results in reasonable calculated free energy profiles. However, the large fluctuations in forces make the autocorrelation function of the force different, and this causes problems when calculating diffusion coefficients.
I have used the same mdp file as in version 4.0.5 with the difference that tau_t is increased from 0.1 to 0.4 (mdp file included below). The simulations are at least 7 ns. Is there anything trivial that I have missed when changing from version 4.0.5 to 4.5.1?
I would really appreciate any help on this issue. Thanks in advance.

Regards,
Emma


The mdp file:

title                                        = x
cpp                                         = /lib/cpp
include                                   =
define                                     =
integrator                            = md
dt                                            = 0.002
nsteps                                   = 7500000
nstlog                    = 25000
nstenergy                            = 30
nstxout                                = 0
nstxtcout                             = 25000
nstvout                                = 0
nstfout                                 = 0
xtc_grps                               = DPPC SOL X
energygrps                          = DPPC SOL X
nstlist                                    = 10
ns_type                               = grid
rlist                                         = 1.0
coulombtype                     = PME
rcoulomb                             = 1.0
vdwtype                              = Cut-off
rvdw                                 = 1.0
tcoupl                                   = Nose-hoover
tc-grps                                  = DPPC SOL X
tau_t                                = 0.4 0.4 0.4
ref_t                                      = 323 323 323
Pcoupl                                  = Parrinello-Rahman
pcoupltype                         = semiisotropic
tau_p                                    = 1.0 1.0
compressibility                  = 4.5e-5 4.5e-5
ref_p                                = 1.0 1.0
gen_vel                                = yes
gen_temp                           = 323
gen_seed                            = 173529
constraints                          = all-bonds

pbc                                         = xyz

optimize_fft                       = yes
unconstrained_start          = no

pull                                         = constraint
pull_geometry                  = cylinder
pull_r1                                  = 1.0
pull_r0                                  = 1.5
pull_group0                        = DPPC
pull_group1                        = X
pull_vec1                             = 0 0 1
pull_init1                             = -0.006
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20120112/7344ace8/attachment.html>