[gmx-users] workaround for PME coulomb energy (fwd)
David L. Bostick
dbostick at physics.unc.edu
Tue Nov 26 19:47:25 CET 2002
Thanks,
Yes.. in ewald_util.c I found that if I turn on debug the charge correction
is written out. There is one catch to the workaround that I haven't
thought of until just now. If in one system, say in my case, a Na ion is
charged (+1) and I turn it off for mdrun -rerun to generate the system
energy, there is a charge correction. Essentially a "background" charge is
introduced that I guess can be thought of as being smeared throughout the
whole system. Doesn't the dipole moment of the system change?
So if I am using the dipole correction term.. this term will change between
the system where the Na charge is +1 and the one where it is 0. I thought
about correcting for this change in the dipole moment when recalculating
the PME energy on the ion, but where should I
put the "background" charge for the case when the Na charge is 0? This is
tricky... I think that maybe the workaround is best used with metallic
boundary conditions. Although, sometimes I can convince myself that it may
be an okay approximation for non-metallic ones. Any thoughts?
An additional question.. as a check.. I am trying to do -rerun while
energy monitoring Na and !Na groups with a large cut-off instead of PME.
This way I can directly get a good estimate of the coulomb energy.
However, I would imagine that Coul-SR:Na-notNa is all I need since I am
using cutoff, but I see that there is a nonzero Coul-LR:Na-notNa. Why is
this?
Thanks,
David
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
David Bostick Office: 262 Venable Hall
Dept. of Physics and Astronomy Phone: (919)962-0165
Program in Molecular and Cellular Biophysics
UNC-Chapel Hill
CB #3255 Phillips Hall dbostick at physics.unc.edu
Chapel Hill, NC 27599 http://www.unc.edu/~dbostick
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
On 26 Nov 2002, David wrote:
> On Tue, 2002-11-26 at 17:40, David L. Bostick wrote:
> >
> >
> > Hello David,
> >
> > Thanks for your help with the workaround. I have an additional question
> > pertaining to your last reply below:
> >
> > > E(lr) = E(lr) (ion-ion) + E(lr) (ion-rest) + E(lr) (rest-rest)
> > > Setting the ion charge to zero will give you E(lr) (rest-rest)
> > > setting the rest to zero will give you E(lr) (ion-ion)
> > > but if you have only one ion that term will be zero. Do check the charge
> > > correction to PME, as that may be different when the total charge
> > > changes.
> >
> > How do I check the charge correction to PME?
> Good question, you'll probably have to look at the equations. (Didn't
> you fid a bug in them at some stage?) The correction is included in the
> PME energy, but if you have two systems with a net charge it may be
> tricky...
>
> >
> > Thanks,
> > David
> >
> > -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
> > David Bostick Office: 262 Venable Hall
> > Dept. of Physics and Astronomy Phone: (919)962-0165
> > Program in Molecular and Cellular Biophysics
> > UNC-Chapel Hill
> > CB #3255 Phillips Hall dbostick at physics.unc.edu
> > Chapel Hill, NC 27599 http://www.unc.edu/~dbostick
> > =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
> >
> > On 23 Oct 2002, David wrote:
> >
> > > On Wed, 2002-10-23 at 17:44, David L. Bostick wrote:
> > > >
> > > > Hello all,
> > > >
> > > > I am interested in using David van der Spoel's method for calculating the
> > > > total coulomb energy of a single ion with the rest of the system. I just
> > > > want to ask a few questions so I can feel comfortable with what I'm doing.
> > > >
> > > > 1)If I take the original .edr file from the simulation, is the total
> > > > coulomb energy of the system Coulomb (SR) + Coulomb (LR) as extracted by
> > > > g_energy? That is.. Coulomb 1-4 doesn't really factor in does it?
> > > In this case it is constant, yes.
> > >
> > > >
> > > > 2)Then if I set the ion charge to zero and rerun to generate a new .edr
> > > > file and calculate the total coulomb energy again as Coulomb (SR) + Coulomb
> > > > (LR), this will be the total PME energy for System_&_!ion-System_&_!ion?
> > > Coulomb (SR) you can do in groups anyway, so you can get that more
> > > accurate from the original run.
> > >
> > > > 3)Then I can subtract the result of #2 from the result of #1 and obtain the
> > > > total coulomb PME energy of System_&_!ion-ion? Since there is only one
> > > > ion, the coulomb energy of ion-ion is zero so I don't have to worry about
> > > > it.
> > > E(lr) = E(lr) (ion-ion) + E(lr) (ion-rest) + E(lr) (rest-rest)
> > > Setting the ion charge to zero will give you E(lr) (rest-rest)
> > > setting the rest to zero will give you E(lr) (ion-ion)
> > > but if you have only one ion that term will be zero. Do check the charge
> > > correction to PME, as that may be different when the total charge
> > > changes.
> > >
> > > > 4)If I write the total ewald sum in the traditional way,
> > > >
> > > > E = E(real) + E(reciprocal) + E(self) + E(dipole_correction)
> > > >
> > > > is the coulomb energy as extracted by g_energy as follows:
> > > >
> > > > Coulomb (SR) = E(real)
> > > >
> > > > and
> > > >
> > > > Coulomb (LR) = E(reciprocal) + E(self) + E(dipole_correction) ?
> > > >
> > > > How exactly are these terms divided among the two quantities Coulomb(SR)
> > > > and Coulomb(LR)?
> > > Coulomb(LR) = E(reciprocal) + E(self) + E(dipole_correction)
> > > Coulomb(SR) = E(real)
> > >
> > >
> > > Groeten, David.
> > > ________________________________________________________________________
> > > Dr. David van der Spoel, Biomedical center, Dept. of Biochemistry
> > > Husargatan 3, Box 576, 75123 Uppsala, Sweden
> > > phone: 46 18 471 4205 fax: 46 18 511 755
> > > spoel at xray.bmc.uu.se spoel at gromacs.org http://zorn.bmc.uu.se/~spoel
> > > ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
> > > _______________________________________________
> > > 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.
> > >
> >
> >
> > _______________________________________________
> > 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.
> --
> Groeten, David.
> ________________________________________________________________________
> Dr. David van der Spoel, Biomedical center, Dept. of Biochemistry
> Husargatan 3, Box 576, 75123 Uppsala, Sweden
> phone: 46 18 471 4205 fax: 46 18 511 755
> spoel at xray.bmc.uu.se spoel at gromacs.org http://zorn.bmc.uu.se/~spoel
> ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
> _______________________________________________
> 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.
>
More information about the gromacs.org_gmx-users
mailing list