[gmx-users] A query about nsteps

[Justin Lemkul]

On 8/20/15 7:49 AM, anu chandra wrote:
> Dear gromcas users,
>
> I carried out a continuation of previous md run for nsteps of 5000000 and
> dt of  0.002 using the following commands,
>
> gmx grompp -f md.mdp -o md2.tpr -c md1.gro -t md1.cpt -r ref.pdb -n
> index.ndx -p topol.top
>
> gmx_mpi mdrun -cpi md1.cpt -deffnm md2
>

This is redundant.  If you restart like this (you should use gmx convert-tpr for 
simple extensions, as grompp has issues preserving pressure coupling 
information), then you don't need -cpi when executing mdrun.

>
> The mdp file following run parameters,
>
> ***************************************************
> integrator              = md
> dt                      = 0.002
> nsteps                  = 5000000
> ;
> nstlog                  = 1000
> nstxout                 = 0
> nstvout                 = 0
> nstfout                 = 0
> nstcalcenergy           = 100
> nstenergy               = 1000
> nstxout-compressed      = 20000
> compressed-x-grps       = System
> ;
> cutoff-scheme           = Verlet
> nstlist                 = 20
> rlist                   = 1.2
> coulombtype             = pme
> rcoulomb                = 1.2
> vdwtype                 = Cut-off
> vdw-modifier            = Force-switch
> rvdw_switch             = 1.0
> rvdw                    = 1.2
> ;
> pbc                     = xyz
> ;
> gen_vel                 = no
> ;
> tcoupl                  = Nose-Hoover
> tc_grps                 = Protein POPC CL_SOL
> tau_t                   = 0.5    0.5     0.5
> ref_t                   = 305.0    305.0    305.0
> ;
> pcoupl                  = Parrinello-Rahman
> pcoupltype              = semiisotropic
> tau_p                   = 5.0
> compressibility         = 4.5e-5   4.5e-5
> ref_p                   = 1.0     1.0
> ;
> constraints             = h-bonds
> constraint_algorithm    = LINCS
> continuation            = yes
> ;
> nstcomm                 = 100
> comm_mode               = linear
> comm_grps               = Protein_POPC CL_SOL
> ;
> refcoord_scaling        = com
> **************************************************************
>
> Though the run completed successfully, I just noticed from the log file
> that the simulation carried out only for 3500000 steps, instead of 5000000
> steps. It looks to me that the simulation started from the previous run
> that completed 1500000 steps and the current run went from the next 3500000
> steps to complete the total run for 5000000 steps. Why this happened? Why
> it counted total run as 5000000 steps rather than continue the current for
> another 5000000 steps?
>

Yes, because of the redundancy above.

> Also, I also noticed that the XTC file is written with 176 frames instead
> of 175 frames (i.e. 3500000/20000)? Is it writing the last step of the
> previous run again?
>

The frame at t=0 is always written to the .xtc, so you always have 
(nsteps/nstxout-compressed)+1 frames.

-Justin

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

Justin A. Lemkul, Ph.D.
Ruth L. Kirschstein NRSA Postdoctoral Fellow

Department of Pharmaceutical Sciences
School of Pharmacy
Health Sciences Facility II, Room 629
University of Maryland, Baltimore
20 Penn St.
Baltimore, MD 21201

jalemkul at outerbanks.umaryland.edu | (410) 706-7441
http://mackerell.umaryland.edu/~jalemkul

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