[gmx-users] heat exchanges

Thu Feb 5 15:59:40 CET 2009

On Feb 5, 2009, at 3:33 PM, Berk Hess wrote:

>
>
> > From: x.periole at rug.nl
> > To: gmx-users at gromacs.org
> > Subject: Re: [gmx-users] heat exchanges
> > Date: Thu, 5 Feb 2009 15:24:28 +0100
> >
> >
> > On Feb 5, 2009, at 3:17 PM, David van der Spoel wrote:
> >
> > > Berk Hess wrote:
> > >> Hi,
> > >> Strictly speaking you should never have thermostats for separate
> > >> parts
> > >> of the system when those parts are coupled through potentials.
> > >> But in practice you can have integration errors which heat or  
> cool
> > >> different
> > >> parts of the system in different ways.
> > >> I think that for most systems only the electrostatics treatment
> > >> will have
> > >> a significant effect. Plain Coulomb cut-off and reaction-field  
> both
> > >> produce
> > >> large integration errors.
> > the problem is due to errors with berendsen bath which eats up the
> > water and therefore the protein gets cooled down while it does not  
> need.
> > The separation avoid this.
> > I think links/shake have similar effects.
>
> No, this problem is not due to the Berendsen thermostat.
you are right, I used a mental short cut.
The  Berendsen coupling does not make mistakes :)) but it corrects them.
>
> The integrations errors in the electrostatics have much more effect
> on the water than on the protein, because the water has higher charges
> and is far more mobile. No thermostat can correct for these errors,
> unless you use multiple groups.
> In addition the Berendsen thermostat will move energy from fast to
> slow motions. This effects becomes much stronger when the thermostat
> has to do a lot of scaling due to the heat generated by cut-off's.
>
> Inaccurate constraining will remove heat from the system, but this
> effect is orders of magnitude smaller than the Coulomb cut-off  
> heating.
ok, good to know.
>
>
> Berk
>
> > >> With PME you should never need multiple t-coupl groups.
> > >> The other issue is slow heat exchange between different parts of
> > >> the system.
> > >> For a protein in solvent I think this is never an issue. But for
> > >> other systems
> > >> it could be. Here it is the question if your simulation is much
> > >> longer
> > >> than this exchange time or not.
> > >> The Berendsen thermostat can produce artifacts when you have  
> weakly
> > >> coupled modes. but with Gromacs 4 there is no reason to use
> > >> Berendsen,
> > >> since we have v-rescale and Nose-Hoover.
> > >> I think you can always use PME, v-rescale or nose-hoover and a  
> single
> > >> t-coupling group.
> > > I think you are correct, but if someone does the test it can
> > > probably be published, if you combine it with recommendations for
> > > an equilibration protocol.
> > >
> > >> Berk
> > >> > Date: Thu, 5 Feb 2009 13:07:21 +0100
> > >> > From: spoel at xray.bmc.uu.se
> > >> > To: gmx-users at gromacs.org
> > >> > Subject: Re: [gmx-users] heat exchanges
> > >> >
> > >> > XAvier Periole wrote:
> > >> > >
> > >> > > On Feb 5, 2009, at 11:02 AM, Carlo Camilloni wrote:
> > >> > >
> > >> > >> Dear Gromacs users,
> > >> > >>
> > >> > >> I am doing some tests with thermostats and I would like to
> > >> know if
> > >> > >> someone has just done it. In particular:
> > >> > > I have not done systematic tests but here are my impressions
> > >> > >>
> > >> > >> 1) How big has to be a protein to be coupled with a separate
> > >> bath?
> > >> > > a few residues ...
> > >> > >>
> > >> > >> 2) Do you know over which time scales a flux of heat is  
> observed
> > >> > >> between protein and solvent using only one temperature  
> group?
> > >> > > few 10-100 ps.
> > >> > This depends a lot on how the rest of the setup is, I would
> > >> think. I had
> > >> > some correspondence on the computational chemistry list about
> > >> this 10
> > >> > years ago, and there people boldly stated that with Nose Hoover
> > >> there is
> > >> > no need for coupling separate groups at all. At the time I  
> could
> > >> not
> > >> > reproduce that with GROMACS, but I did not persevere it. It  
> could
> > >> also
> > >> > have to do with flexibility, and/or using united atoms without
> > >> charges,
> > >> > i.e. that there is little energetic coupling between e.g. water
> > >> and a
> > >> > hydrophobic substance.
> > >> >
> > >> > It would be worthwhile to do systematic tests with different
> > >> thermostats
> > >> > and coupling schemes. If it turns out that the same problems
> > >> exist in
> > >> > Nose Hoover thermostats then it is probably publishable as  
> well.
> > >> >
> > >> >
> > >> >
> > >> > >>
> > >> > >>
> > >> > >> Thanks,
> > >> > >>
> > >> > >> Carlo Camilloni
> > >> > >>
> > >> > >> --
> > >> > >> Dr Carlo Camilloni
> > >> > >> Department of Physics, University of Milano
> > >> > >> via Celoria, 16 - 20133 Milano, Italy
> > >> > >> phone: +39-02-50317654
> > >> > >> carlo.camilloni at mi.infn.it
> > >> > >> http://merlino.mi.infn.it/~carlo
> > >> > >>
> > >> > >>
> > >> > >>
> > >> > >>
> > >> > >>
> > >> > >> _______________________________________________
> > >> > >> gmx-users mailing list gmx-users at gromacs.org
> > >> > >> http://www.gromacs.org/mailman/listinfo/gmx-users
> > >> > >> Please search the archive at http://www.gromacs.org/search
> > >> before
> > >> > >> posting!
> > >> > >> Please don't post (un)subscribe requests to the list. Use  
> thewww
> > >> > >> interface or send it to gmx-users-request at gromacs.org.
> > >> > >> Can't post? Read http://www.gromacs.org/mailing_lists/users.php
> > >> > >
> > >> > > _______________________________________________
> > >> > > gmx-users mailing list gmx-users at gromacs.org
> > >> > > http://www.gromacs.org/mailman/listinfo/gmx-users
> > >> > > Please search the archive at http://www.gromacs.org/search
> > >> before posting!
> > >> > > Please don't post (un)subscribe requests to the list. Use  
> thewww
> > >> > > interface or send it to gmx-users-request at gromacs.org.
> > >> > > Can't post? Read http://www.gromacs.org/mailing_lists/users.php
> > >> >
> > >> >
> > >> > --
> > >> > David van der Spoel, Ph.D., Professor of Biology
> > >> > Molec. Biophys. group, Dept. of Cell & Molec. Biol., Uppsala
> > >> University.
> > >> > Box 596, 75124 Uppsala, Sweden. Phone: +46184714205. Fax:
> > >> +4618511755.
> > >> > spoel at xray.bmc.uu.se spoel at gromacs.org http://folding.bmc.uu.se
> > >> > _______________________________________________
> > >> > gmx-users mailing list gmx-users at gromacs.org
> > >> > http://www.gromacs.org/mailman/listinfo/gmx-users
> > >> > Please search the archive at http://www.gromacs.org/search  
> before
> > >> posting!
> > >> > Please don't post (un)subscribe requests to the list. Use the
> > >> > www interface or send it to gmx-users-request at gromacs.org.
> > >> > Can't post? Read http://www.gromacs.org/mailing_lists/users.php
> > >>  
> ------------------------------------------------------------------------
> > >> Express yourself instantly with MSN Messenger! MSN Messenger <http://clk.atdmt.com/AVE/go/onm00200471ave/direct/01/
> > >> >
> > >>  
> ------------------------------------------------------------------------
> > >> _______________________________________________
> > >> gmx-users mailing list gmx-users at gromacs.org
> > >> http://www.gromacs.org/mailman/listinfo/gmx-users
> > >> Please search the archive at http://www.gromacs.org/search before
> > >> posting!
> > >> Please don't post (un)subscribe requests to the list. Use the www
> > >> interface or send it to gmx-users-request at gromacs.org.
> > >> Can't post? Read http://www.gromacs.org/mailing_lists/users.php
> > >
> > >
> > > --
> > > David van der Spoel, Ph.D., Professor of Biology
> > > Molec. Biophys. group, Dept. of Cell & Molec. Biol., Uppsala
> > > University.
> > > Box 596, 75124 Uppsala, Sweden. Phone:	+46184714205. Fax:  
> +4618511755.
> > > spoel at xray.bmc.uu.se	spoel at gromacs.org http://folding.bmc.uu.se
> > > _______________________________________________
> > > gmx-users mailing list gmx-users at gromacs.org
> > > http://www.gromacs.org/mailman/listinfo/gmx-users
> > > Please search the archive at http://www.gromacs.org/search before
> > > posting!
> > > Please don't post (un)subscribe requests to the list. Use the www
> > > interface or send it to gmx-users-request at gromacs.org.
> > > Can't post? Read http://www.gromacs.org/mailing_lists/users.php
> >
> > _______________________________________________
> > gmx-users mailing list gmx-users at gromacs.org
> > http://www.gromacs.org/mailman/listinfo/gmx-users
> > Please search the archive at http://www.gromacs.org/search before  
> posting!
> > Please don't post (un)subscribe requests to the list. Use the
> > www interface or send it to gmx-users-request at gromacs.org.
> > Can't post? Read http://www.gromacs.org/mailing_lists/users.php
>
> Express yourself instantly with MSN Messenger! MSN Messenger  
> _______________________________________________
> gmx-users mailing list    gmx-users at gromacs.org
> http://www.gromacs.org/mailman/listinfo/gmx-users
> Please search the archive at http://www.gromacs.org/search before  
> posting!
> Please don't post (un)subscribe requests to the list. Use the
> www interface or send it to gmx-users-request at gromacs.org.
> Can't post? Read http://www.gromacs.org/mailing_lists/users.php

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20090205/8fe1cf96/attachment.html>