[gmx-users] how to switch from NPT to NVT ensemble during equilibration

[Mark Abraham]

On 10/02/2011 10:54 AM, Rini Gupta wrote:
> Dear gmx users,
>
> I am using gromacs (version 4.0.7) first time
> to setup a 2-butoxyethanol-water simulation.
> I created topology and coordinate file for BE using AUTOMATED TOPOLOGY 
> BUILDER server.
> It created a topology file (for united atom) compatible with GROMOS 
> ffG53a6 forcefield.
> Then a generated a box containing 20 BE and 480 water molecules using
> genbox.
>
> When I performed energy minimization followed by mdrun using NPT ensemble.
> I get
>
> Potential Energy = -1.56616474544600e+04
> Maximum force = 9.72854664927673e+02 on atom 910
>
>
> Then, I run for 200ps equlibration using NPT ensemble with Berendsen
> thermostat and P coupling
> I want to use NVT ensemble for my calculations, so it is ok to switch 
> from NPT to NVT ensemble during equilibration?

Sure. See 
http://www.gromacs.org/Documentation/How-tos/Steps_to_Perform_a_Simulation

> First, I did equilibation for 200ps using NPT ensemble
> I get
> Total Energy Temperature Pressure (bar) Cons. rmsd ()
> -1.81617e+04 3.00167e+02 -4.32808e+00 0.00000e+00
> and,
> My box length changes from 2.7 nm to 2.6503 nm
>
> Then using state.cpt with mdrun I run for another 200ps using NVT 
> settings in .mpd file ( I turn off the P-coupling).
>
> I use the following commands:
> I regenerated new .tpr file
> grompp_d_mpi -f grompp.mdp -c confout.gro -p topol.top -o topol.tpr
>
> mdrun_d_mpi -s topol.tpr -cpi state.cpt
>
> Simulation is completed with warning
>
> WARNING: The checkpoint state entries do not match the simulation,
> see the log file for details

OK and what did you learn from the .log file about this? (Probably, it 
restarted from the coordinates of the confout.gro, which loses precision 
and some of the value of your initial equilibration)

> and Pressure again increases up to 60 bar

No, it probably doesn't. See 
http://www.gromacs.org/Documentation/Terminology/Pressure

>
> Total Energy Temperature Pressure (bar) Cons. rmsd ()
> -1.82304e+04 2.99992e+02 6.20706e+01 0.00000e+00
>
>
> I want to know if this is a right procedure for switching ensemble.
> do i need to generate new velocities during second run using NVT ensemble?
> If this is correct, how can i reach the target of Pressure 1 bar using 
> this approach?

See 
http://www.gromacs.org/Documentation/How-tos/Extending_Simulations#Version_4

> I am using following topology file:
> Can anyone please tell me if this topology is o.k to use

It's syntactically correct because grompp doesn't complain. Whether its 
a sensible model physics can sometimes be rejected off-the-cuff, but 
can't really be confirmed without actually testing against some other data.

Mark

> ; Include forcefield parameters
> #include "ffG53a6.itp"
>
> [ moleculetype ]
> ; Name nrexcl
> G269 3
> [ atoms ]
> ; nr type resnr resid atom cgnr charge mass total_charge
> 1 OE 1 G269 O 1 -0.345 15.9994
> 2 CH2 1 G269 C 1 0.151 14.0270
> 3 CH2 1 G269 C 1 0.194 14.0270 ; 0.000
> 4 CH2 1 G269 C 2 0.231 14.0270
> 5 OA 1 G269 O 2 -0.617 15.9994
> 6 H 1 G269 H 2 0.386 1.0080 ; 0.000
> 7 CH2 1 G269 C 3 -0.035 14.0270
> 8 CH2 1 G269 C 3 0.143 14.0270
> 9 CH3 1 G269 C 3 -0.108 15.0350 ; -0.000
> ; total charge of the molecule: 0.000
> [ bonds ]
> ; ai aj funct c0 c1
> 1 2 2 0.1430 8.1800e+06
> 1 3 2 0.1430 8.1800e+06
> 2 4 2 0.1520 5.4300e+06
> 3 7 2 0.1520 5.4300e+06
> 4 5 2 0.1430 8.1800e+06
> 5 6 2 0.1000 2.3200e+07
> 7 8 2 0.1530 7.1500e+06
> 8 9 2 0.1530 7.1500e+06
> [ pairs ]
> ; ai aj funct ; all 1-4 pairs but the ones excluded in GROMOS itp
> 1 5 1
> 1 8 1
> 2 6 1
> 2 7 1
> 3 4 1
> 3 9 1
> [ angles ]
> ; ai aj ak funct angle fc
> 2 1 3 2 109.50 380.00
> 1 2 4 2 109.50 320.00
> 1 3 7 2 109.50 320.00
> 2 4 5 2 111.00 530.00
> 4 5 6 2 108.53 443.00
> 3 7 8 2 111.00 530.00
> 7 8 9 2 111.00 530.00
> [ dihedrals ]
> ; GROMOS improper dihedrals
> ; ai aj ak al funct angle fc
> [ dihedrals ]
> ; ai aj ak al funct ph0 cp mult
> 3 1 2 4 1 0.00 1.26 3
> 2 1 3 7 1 0.00 1.26 3
> 1 2 4 5 1 0.00 2.53 3
> 1 3 7 8 1 0.00 3.77 3
> 2 4 5 6 1 0.00 1.26 3
> 3 7 8 9 1 0.00 3.77 3
> [ exclusions ]
> ; ai aj funct ; GROMOS 1-4 exclusions
>
>
>
>
> ; Include water topology
> #include "spce.itp"
> [ system ]
> ; Name
> BE in Water
>
> [ molecules ]
> ; Compound #mols
> G269 20
> SOL 480
>
> Please help me in this regard.
>
> Thanks and Regards,
> Rini
>
> ----------------
> Dr. Rini Gupta
> Postdoctoral Fellow
> University of British Columbia
> Vancouver
>
>
>
> <http://sigads.rediff.com/RealMedia/ads/click_nx.ads/www.rediffmail.com/signatureline.htm@Middle?> 
>
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20110210/aff4fe6a/attachment.html>