[gmx-users] Errors in constraint pulling

[Hideya Nakamura]

Dear GROMACS users,

I am trying to calculate the potential of mean force (PMF) of a
molecule (A) across phospholipid bilayer (LIP) using constraint
method.
I am using GROMACS ver 4.0.4.
RUN parameters are as follows
-------------------------------------------------------------------------------
integrator               = md
tinit                    = 0
dt                       = 0.002 ;[ps]
nstlist                  = 10 ;[step]
ns_type                  = grid
rlist                    = 1.0
pbc                      = xyz

coulombtype              = pme
rcoulomb                 = 1.0	;[nm]
epsilon-r                = 1    ;(1)
vdw-type                 = switch
rvdw-switch              = 0.9 ;[nm]
rvdw                     = 1.0 ;[nm]
DispCorr                  = EnerPres

fourierspacing		 = 0.12 ;[nm] (0.12)
fourier_nx               = 0       ;(0)
fourier_ny               = 0       ;(0)
fourier_nz               = 0       ;(0)
pme_order                = 4       ;(4)
ewald_rtol               = 1e-05   ;(1e-5)
epsilon_surface          = 0       ;(0)
optimize_fft             = no

tcoupl			 = berendsen
tc_grps                  = LIP SOL A
tau_t                    = 0.1 0.1 0.1 ;[ps]
ref_t                    = 300 300 300 ;[K]
Pcoupl                   = berendsen
Pcoupltype               = semiisotropic   ;useful for membrane	
tau_p                    = 1.0 1.0         ;[ps]
compressibility          = 4.5e-05 4.5e-05 ;[bar^-1]
ref_p                    = 1.0 1.0         ;[bar]

gen_vel                  = no ;yes or no
gen_temp                 = 300 ;[K]
gen_seed                 = 113 ;(173529)

constraints              = all-bonds  ;all-bonds/none
constraint-algorithm     = LINCS
unconstrained-start      = no
lincs-order              = 4   ;(4)
lincs-warnangle          = 30  ;(30)

; COM PULLING
pull                     = constraint
pull_geometry            = direction
pull_dim                 = N N Y
pullstart                = no
pull_nstxout             = 10
pull_nstfout             = 10
pull_ngroups             = 1
pull_group0              = LIP
pull_group1              = A
pull_vec1                = 0.0 0.0 -1.0
pull_init1               = -3.50
pull_rate1               = 0
pull_k1                  = 500
-------------------------------------------------------------------------------
When I simulated under above parameters, following warnings and errors
were generated.

-Warning: pressure scaling more than 1%
-LINCS warning:

These errors were always generated when a molecule (pull_gropu1) was
existing in water phase.
After simulation, serious disruption of a molecule (pull_gropu1) was observed.

However, when a molecule (pull_gropu1) was existing inside lipid
membrane, no error was observed and calculation was successfully
conducted in constrained simulation.

I have done long time (5 nsec) simulations before constrained
simulation using umbrella method, thus I believe the initial condition
is fully equilibrated.

Could some one please give me an advice to calculate PMF?

Sincerely


___________________________________________________________________
Hideya Nakamura, Ph.D.

Post Doctoral Associate
University of Florida, Particle Engineering Research Center (PERC)
mail to: hideyanakamura at gmail.com