[gmx-users] Issues with NEMD simulations

[Justin A. Lemkul]

shikha nangia wrote:
> Dear all,
> 
> 1. I am performing MD simulation of a MARTINI coarse grained polymeric 
> chains (named NGL in my mdp file below) in Coarse grained water (named W 
> in mdp file below).
> 
> 2. My equilibration and MD runs have been running perfectly until now. 
> I  have started doing NEMD simulations to get viscosity. The starting 
> structure for NEMD  came from an equilibrated system.
> 
> 3. I am pretty sure that the error is in the mdp file. I am new to NEMD 
> so it will be great if I can get some pointers of solving the issue. Is 
> there a NEMD gromacs tutorial on the web?
> 

Not likely.  If its not posted on http://www.gromacs.org/Documentation/Tutorials 
and you can't find it with Google, then it probably doesn't exist, unfortunately.

> Here is the full mdp file for the NEMD run:
> title                    = Martini
> cpp                      = /usr/bin/cpp
> integrator               = md
> tinit                    = 0.0
> dt                       = 0.01
> nsteps                   = 500
> nstcomm                  = 1
> comm-grps                         = system
> nstxout                  = 10
> nstvout                  = 10
> nstfout                  = 10
> nstlog                   = 10
> nstenergy                = 10
> nstxtcout                = 2500
> xtc_precision            = 100
> xtc-grps                 =
> energygrps               = system
> nstlist                  = 1
> ns_type                  = grid
> pbc                      = xyz
> rlist                    = 1.7
> coulombtype              = Shift
> rcoulomb_switch          = 0.0
> rcoulomb                 = 1.2
> epsilon_r                = 15
> vdw_type                 = Shift
> rvdw_switch              = 0.9
> rvdw                     = 1.2
> DispCorr                 = EnerPres
> tcoupl                   = V-rescale
> tc-grps                  = NGL W
> tau_t                    = 1.0 1.0
> ref_t                    = 300. 300.
> Pcoupl                   = No
> Pcoupltype               = Isotropic
> tau_p                    = 1.0
> compressibility          = 1e-5
> ref_p                    = 1.0
> gen_vel                  = Yes

Well, you started with an equilibrated system, but then re-generating velocities 
destroys that.

> gen_temp                 = 300
> gen_seed                 = 473529
> constraints              = None
> constraint_algorithm     = Lincs
> unconstrained_start      = no
> lincs_order              = 1
> lincs_warnangle          = 30
> 
> ; Non-equilibrium MD stuff
> acc-grps                 =  NGL W
> accelerate               =  0.1 0 0 -0.1 0 0

Are NGL and W mixed?  Or are they separate phases?  If they're mixed and being 
pushed in opposite directions, you're just smashing your system together.  Even 
if they're layered, you may be getting nasty collisions at the interface of the 
two phases.  You can investigate the source of the LINCS errors you posted 
before by following the tips here:

http://www.gromacs.org/Documentation/Terminology/Blowing_Up#Diagnosing_an_Unstable_System

Based on the fact that you're saving every 10 steps, you should be able to get a 
pretty good look at what's going on, and in conjunction with the atom numbers 
that the LINCS warnings print, you can pinpoint exactly where everything starts 
to break down.

-Justin

> freezegrps               =  
> freezedim                =  
> cos_acceleration         = 0.1
> deform                   = 
> 
> Best,
> SN
> Dr. Shikha Nangia
> 201D Chemistry Building
> Department of Chemistry
> The Pennsylvania State University
> University Park, PA 16802 
> Phone(o): 814-863-0289
> Phone(cell): 612-599-9449
> 

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

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

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