[gmx-users] pathologically expanding box

[Justin A. Lemkul]

Greg Bowman wrote:
> Hello,
> 
> Has anyone else experienced a pathologically expanding box during 
> equilibration in the NPT ensemble?  I've solvated my system with 
> editconf/genbox, energy minimized, equilibrated in NVT with the protein 
> coordinates restrained, and then equilibrated in NPT without any 
> position restraints.  During the NPT equilibration all the box 
> dimensions are doubling and the density decreases drastically.
> 
> I'm using the v-rescale thermostat and the Berendsen barostat in gromacs 
> 4.5.3 with the AMBER03 force field and tip3p water.  I've also tried 
> OPLSAA with SPC water, a number of different box types (cubic, 
> octahedral, dodecahedral), and shorter tau_t and tau_p settings.  I also 
> tried adding another energy minimzation step between my NVT and NPT phases.
> 

Is the temperature, potential energy, etc. stable after NVT?

Is the increase in box size steady, oscillating, or sudden?

Does the problem persist with different combinations of thermostat and barostat, 
or is it limited to V-rescale + Berendsen?

-Justin

> My mdp settings for the NPT ensemble are below.
> 
> Thanks for your help.
> 
> Greg
> 
> ; RUN CONTROL PARAMETERS
> ;define              = -DPOSRES
> integrator          = md
> tinit               = 0
> nsteps              = 50000
> dt                  = 0.002000
> comm-mode           = linear
> nstcomm             = 4
> 
> ; CONSTRAINTS
> constraints         = all-bonds
> lincs_order         = 6
> lincs_iter          = 2
> 
> ; NEIGHBORSEARCHING PARAMETERS
> ns_type             = grid
> nstlist             = 4
> rlist               = 1.2
> pbc                 = xyz
> 
> ; ELECTROSTATICS
> coulombtype         = pme ;-switch
> rcoulomb            = 1.2
> ;rcoulomb-switch     = 0.800000
> fourierspacing      = 0.120000
> fourier_nx          = 0
> fourier_ny          = 0
> fourier_nz          = 0
> pme_order           = 6
> ewald_rtol          = 0.000001
> ewald_geometry      = 3d
> epsilon_surface     = 0
> optimize_fft        = yes
> epsilon_r           = 80
> 
> ; VDW
> vdw-type            = switch
> rvdw                = 1.2
> rvdw-switch         = 1.1
> DispCorr            = EnerPres
> 
> ; OUTPUT CONTROL OPTIONS
> ; Output frequency for coords (x), velocities (v) and forces (f) =
> nstxout             = 0
> nstvout             = 0
> nstfout             = 0
> nstlog              = 0
> energygrps          = System
> nstenergy           = 100
> nstcalcenergy       = 1
> nstxtcout           = 0
> xtc-grps            = Protein
> 
> Tcoupl              = v-rescale
> tau_t               = 0.5
> tc-grps             = System
> ref_t               = 300
> ld_seed             = -1
> 
> Pcoupl              = berendsen
> Pcoupltype          = isotropic
> tau_p               = 5
> compressibility     = 4.5e-5
> ref_p               = 1.0
> 
> gen_vel             = yes
> gen_temp            = 300.0
> gen_seed            = -1
> 
> 

-- 
========================================

Justin A. Lemkul
Ph.D. Candidate
ICTAS Doctoral Scholar
MILES-IGERT Trainee
Department of Biochemistry
Virginia Tech
Blacksburg, VA
jalemkul[at]vt.edu | (540) 231-9080
http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin

========================================