[gmx-users] What is the difference between joined and seperate tem coupling?

[Tanping Li]

Dear Tsjerk,

Thank you very much for your reply and thinking. I
calculated the Tem from the g_traj, in which I need to
give 1.5 factor of Tem in order to get the final Tem.
This is what I reported. For the for the ions, I also
gave it the factor 1.5. Maybe this is the origin that
the Tem of ions is so crazy? Should I time 1.5 to
ions? 
Thank you again.


Best
Tanping


--- Tsjerk Wassenaar <tsjerkw at gmail.com> wrote:

> Hi Tanping Li,
> 
> What this tells you is that you're independently
> coupling your sodium ions
> to the heat bath, which makes their temperature
> fluctuate like crazy and may
> introduce artefacts. Couple the sodium ions together
> with the solvent. This
> has been dealt with on the user list several times,
> check the archives.
> Also, it's a bit dubious to talk about the
> temperature of a single
> residue... Temperature is a collective property.
> Don't start to seperately
> couple a single residue. It's ok like that.
> 
> Cheers,
> 
> Tsjerk
> 
> On 11/1/05, Tanping Li <jia_11_osu at yahoo.com> wrote:
> >
> > Dear Devid,
> >
> > Under pme and globlly T-coupling at 295K, I get
> the
> > tem for protein, water and Na in my system like
> this:
> >
> > Protein Sol Na
> > Average 293 297 434
> >
> > Is it acceptable?
> >
> >
> > The tem of Na is very large. Does that tells me
> that
> > thfe charged groups in protein will also deviate
> far
> > away fro the average Tem? I am analysising the
> > interaction between a specific residue and close
> > waters. Do I need to seperately couple their tem
> since
> > I find out that tem of this specific residue(trp)
> is a
> > little bit from 295K. Thank you very much for your
> > help.
> >
> >
> > Tanping
> >
> >
> >
> > --- David <spoel at xray.bmc.uu.se> wrote:
> >
> > > On Tue, 2005-11-01 at 08:10 -0800, Tanping Li
> wrote:
> > > > Devid,
> > > >
> > > > Thank you very much for your reply. I used
> globlly
> > > > T-coupling and pme. Now I am nervous about it.
> > > > Firstly, I want to check the temprature of
> protein
> > > and
> > > > Sol seperately. Is there any way in Gromacs to
> get
> > > the
> > > > Tem of specific group, instead of calculating
> Tem
> > > by
> > > > the velocity by myself?
> > > >
> > >
> > > g_traj
> > >
> > > >
> > > > Tanping
> > > >
> > > > --- David <spoel at xray.bmc.uu.se> wrote:
> > > >
> > > > > On Tue, 2005-11-01 at 08:16 +0100, Erik
> Lindahl
> > > > > wrote:
> > > > > > Hi,
> > > > > >
> > > > > > If you use PME you're fine with either. If
> you
> > > > > have cut-offs and
> > > > > > large drifts in energy you run a risk of
> e.g.
> > > the
> > > > > water beeing cooler
> > > > > > than the protein if you only use global
> > > coupling.
> > > > >
> > > > > I'm not sure this is correct. Heat exchange
> > > between
> > > > > components for which
> > > > > the forces are relatively weak is as far as
> I
> > > know a
> > > > > problem. Please
> > > > > test it and report back to the list. It may
> also
> > > be
> > > > > slightly worse for
> > > > > GROMOS than for OPLS since OPLS has more
> charged
> > > > > atoms in the protein
> > > > > than GROMOS.
> > > > >
> > > > > Why don't you do four 500 ps sims with
> GROMOS
> > > and
> > > > > OPLS and with separate
> > > > > coupling groups and only one, and make a
> table
> > > of
> > > > > the protein
> > > > > temperatures in respective simulations. You
> will
> > > > > have to save velocities
> > > > > relatively often to compute the temperature
> in
> > > the
> > > > > simulation where you
> > > > > only have one group.
> > > > >
> > > > > >
> > > > > > Cheers,
> > > > > >
> > > > > > Erik
> > > > > >
> > > > > >
> > > > > > On Nov 1, 2005, at 1:28 AM, Tanping Li
> wrote:
> > > > > >
> > > > > > > Dear all,
> > > > > > >
> > > > > > > I am simulating the hyrdation issume of
> > > protein
> > > > > under
> > > > > > > NVT condition. We use Gromos96 force
> field.
> > > I
> > > > > noticed
> > > > > > > that in the tutorial mdp file, the
> protein
> > > and
> > > > > solvent
> > > > > > > are tem coupled seperately. Could I use
> the
> > > > > global
> > > > > > > tem-coupling? Is there any real
> difference
> > > > > between
> > > > > > > them?
> > > > > > > Thank you very much for your reply.
> > > > > > >
> > > > > > >
> > > > > > > Best
> > > > > > > Tanping Li
> > > > > > >
> > > _______________________________________________
> > > > > > > gmx-users mailing list
> > > > > > > gmx-users at gromacs.org
> > > > > > >
> > > > >
> > >
> http://www.gromacs.org/mailman/listinfo/gmx-users
> > > > > > > Please don't post (un)subscribe requests
> to
> > > the
> > > > > list. Use the
> > > > > > > www interface or send it to
> > > > > gmx-users-request at gromacs.org.
> > > > > > >
> > > > > > >
> > > > > >
> > > > > >
> > > > >
> > > >
> > >
> >
>
-----------------------------------------------------------
> > > > > > Erik Lindahl <lindahl at sbc.su.se>
> > > > > > Assistant Professor, Stockholm
> Bioinformatics
> > > > > Center
> > > > > > Stockholm University, SE 106 91 Stockholm
> > > > > > Phone: +46 8 5537 8564 Fax: +46 8 5537
> > > 8214
> > > > > >
> > > > > >
> > > > > >
> > > > > >
> > > > > >
> > > _______________________________________________
> > > > > > gmx-users mailing list
> > > > > > gmx-users at gromacs.org
> 
=== message truncated ===>
_______________________________________________
> gmx-users mailing list
> gmx-users at gromacs.org
> http://www.gromacs.org/mailman/listinfo/gmx-users
> Please don't post (un)subscribe requests to the
> list. Use the 
> www interface or send it to
gmx-users-request at gromacs.org.