[gmx-users] Decouple

[Justin Lemkul]

Please keep the discussion on the gmx-users mailing list; I am not a private 
tutor.  Comments embedded below.

On 6/20/13 8:35 PM, Hari Pandey wrote:
> Hi Justin,
>   Thanks for last help.
>
> I have following (NVT.mdp) (NPT.mdp)  and (NVE.mdp)   .  I have three parts of
> system, A, B and C.
> I want to put thermostat on A and C and couple the tempereture and do not put
> any thermostat on B.
>
> What I want is: after some ps,  temperature of A and C should be constant (i.e
> reach up to steady state) and  B may not be.
>
> for that My NVT.mdp is:
> ; simulation at 300K and 2 ps is on
>          constraints =all-bonds
>          integrator      =md
>          dt              =0.001 ; ps
>          nsteps          =100000 ; total 100 ps
>          nstcomm         =10
>          nstxout         =1000
>          nstxtcout       =0
>          nstvout         =0
>          nstfout         =0
> nstenergy       =100
>          nstlist         =100

This value of nstlist is much too large.

>          ns_type         =grid
>          rlist           =0.5
>          coulombtype     =pme
>          rcoulomb        =0.5
>          vdwtype         =cut-off
>          rvdw            =0.5

Unless you are using some custom force field that requires 0.5-nm cutoffs, these 
values are incorrect.

>          pme_order       =4
> ewald_rtol      =1e-5
>          optimize_fft    =yes
>          DispCorr        =no
>
> ;Brendsen tempereture coupling is on
>          Tcoupl          = nose-hoover

Nose-Hoover is a poor choice for initial equilibration, especially when 
generating velocities.

>          tau_t           =0.001  -1   0.001

Setting tau_t = dt is a bad idea and grompp should have warned you about this.

>          tc-grps         =A  B   C
>
>          ref_t =750      300       350
>
>
> ;pressure coupling is on
>          Pcoupl          =no
>          Pcoupltype      =isotropic
>          tau_p           =0.5
>          compressibility =1e-5
>          ref_p           =0.5
> ;generate velocities at 300 k i.e. at room tempereture
>          gen_vel         =yes
>          gen_temp        =750  300  350
> gen_seed        =-1
>
> MY NPT.mdp is:
>
> ( here all output control parameters also)
>
> ;Brendsen tempereture coupling is on
> Tcoupl          =nose-hoover
>          tau_t           =1  -1   1
>          tc-grps         =NCALPHA MIDDLE NCNN
>          ref_t           =750  300 350
>
>
>
> ;pressure coupling is on
>          Pcoupl          =Berendsen
>          Pcoupltype      =isotropic
>          tau_p           =0.5
>          compressibility =1e-5
> ref_p           =1
> ;generate velocities at 300 k i.e. at room tempereture
>          gen_vel         =no
>          gen_temp        =750 300 350
>          gen_seed        =-1
>
> MY NVE.mdp is:
> ( here all output control parameters also)
> tc-grps         = A  B  C
>          ref_t           =750 300 300
>          energygrps      = NCALPHA  MIDDLE  NCNN
>          tcoupl         = nose-hoover
>          tau-t          = 1  -1   1

If you're using thermostats, you're not simulating in an NVE ensemble.

> ;pressure coupling is on
>          Pcoupl          =no
>          ;Pcoupltype      =isotropic
>          ;tau_p =0.5
>          ;compressibility =1e-5
>          ;ref_p           =0.5
> ;generate velocities at 300 k i.e. at room tempereture
>          gen_vel         =no
>        ;  gen_temp        =750  300  350
>         ; gen_seed        =-1
>
>
> What I did is:
>
> pdb2gmx - argnew.pdb -o fws.pdb -p  fws.top;
> editconf -f fws.pdb -bt dodecahedron -o fws.pdb -d 1.0;
> grompp-f em.mdp -c fws.pdb -p fws.top -n index.ndx -o em.tpr -maxwarn 5;
> mdrun -deffnm  em -v;
> grompp -f nvt.mdp -c em.gro -p fws.top -n index.ndx  -o nvt.tpr -maxwarn 5;
> mdrun -deffnm nvt -v;
> grompp -f npt.mdp -c nvt.gro -p fws.top -n index.ndx  -o npt.tpr -maxwarn 5;
> mdrun -deffnm npt -v
> grompp -f nve.mdp -c npt.gro -p fws.top -n index.ndx  -o nve.tpr -maxwarn 5;
> mdrun -deffnm nve -v;
>
> g_energy -f nve.edr -s nve.tpr -o F1.xvg
>
>
> But the system do not get equilibrated and A, B has not steady state
> temperature  after   time
>   even 100 ps. please help me, where I did wrong
>

You have numerous issues, listed above.

-Justin

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

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

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