[gmx-users] cool down problem

[Riccardo Concu]

Dear Justin,
I ran a short simulation (200ps) with the cool down under NVT and now it
seems the system is working. In addition, I removed from the NVT options
the tc for all the groups.
Thank you,
Riccardo
El jue, 31-10-2013 a las 08:31 -0400, Justin Lemkul escribió:
> 
> On 10/31/13 8:25 AM, Riccardo Concu wrote:
> > Dear all, I'm having some problems during the cool-down process after
> > the NVT.
> > I'm running the NVT process at a T of 500K for 3ns and everything is
> > going well but when I cool down the system blows up. I am running the
> > cool down for 100ps, I tried to run the process for longer time (500 ps)
> > and decrease nstlist option up to 5 but it seems it doesn't work. The EM
> > seems to be fine with very low forces:
> > Potential Energy  = -3.1736706e+05
> > Maximum force     =  2.6573439e+01 on atom 105
> > Norm of force     =  3.1027287e-01
> >
> > The NVT is compiled with the following mdp:
> > define      = -DPOSRES  ; position restrain the protein and ligand
> > ; Run parameters
> > integrator  = md        ; leap-frog integrator
> > nsteps      = 1500000     ; 2 * 50000 = 100 ps
> > dt          = 0.002     ; 2 fs
> > ; Output control
> > nstxout     = 10000       ; save coordinates every 0.2 ps
> > nstvout     = 10000       ; save velocities every 0.2 ps
> > nstenergy   = 10000       ; save energies every 0.2 ps
> > nstlog      = 10000       ; update log file every 0.2 ps
> > energygrps  = NAP NA WAT MET SI3 SI- I LI+ LI-
> > ; Bond parameters
> > continuation    = no            ; first dynamics run
> > constraint_algorithm = lincs    ; holonomic constraints
> > constraints     = all-bonds     ; all bonds (even heavy atom-H bonds)
> > constrained
> > lincs_iter      = 1             ; accuracy of LINCS
> > lincs_order     = 4             ; also related to accuracy
> > ; Neighborsearching
> > vdwtype = cutoff
> > ns_type     = grid      ; search neighboring grid cells
> > nstlist     = 20         ; 10 fs
> > rlist       = 1.0       ; short-range neighborlist cutoff (in nm)
> > rcoulomb    = 1.0       ; short-range electrostatic cutoff (in nm)
> > rvdw        = 1.0       ; short-range van der Waals cutoff (in nm)
> > ; Electrostatics
> > coulombtype     = PME       ; Particle Mesh Ewald for long-range
> > electrostatics
> > pme_order       = 4         ; cubic interpolation
> > fourierspacing  = 0.16      ; grid spacing for FFT
> > ; Temperature coupling
> > tcoupl      = V-rescale                     ; modified Berendsen
> > thermostat
> > tc-grps     = NAP NA WAT MET SI3 SI- I LI+ LI-          ; two coupling
> > groups - more accurate
> 
> Frankly, I'm amazed this was stable.  I can't imagine that any system justifies 
> this many individual thermostats.
> 
> <snip>
> 
> > Pcoupl                   = Parrinello-Rahman
> 
> I doubt that you want to try to use Parrinello-Rahman while simultaneously 
> decreasing the temperature.  Even for equilibration runs at stable temperature, 
> P-R is a poor choice.  I would suggest trying your cooling under NVT, or if you 
> absolutely need to do it under NPT, use Berendsen instead; it is much more 
> forgiving.
> 
> -Justin
> 
> -- 
> ==================================================
> 
> Justin A. Lemkul, Ph.D.
> Postdoctoral Fellow
> 
> Department of Pharmaceutical Sciences
> School of Pharmacy
> Health Sciences Facility II, Room 601
> University of Maryland, Baltimore
> 20 Penn St.
> Baltimore, MD 21201
> 
> jalemkul at outerbanks.umaryland.edu | (410) 706-7441
> 
> ==================================================