[gmx-users] dummy atoms for amber free energy

[Berk Hess]

>From: Maik Goette <mgoette at mpi-bpc.mpg.de>
>Reply-To: Discussion list for GROMACS users <gmx-users at gromacs.org>
>To: Discussion list for GROMACS users <gmx-users at gromacs.org>
>Subject: Re: [gmx-users] dummy atoms for amber free energy
>Date: Tue, 07 Feb 2006 10:56:46 +0100
>
>Hi David
>
>That's also a point of discussion in our group.
>My personal opinion is, that a mass, which is connected via e.g. the bond 
>potential would fluctuate in the harmonic function and those energy terms 
>should cancel out over time. For the dummy-masses have no other 
>interactions, there won't occur any frictional energy contribution.
>So everything should be fine.
>Anyway, grompp don't like 0 masses. This may be, because F/m=a and this 
>would lead to a division with 0, which isn't defined. I never tried to 
>compare a run with full dummy masses and masses set to small values, e.g. 
>0.001.
>
>For this is very intuitive, I would be very thankful for every hint to 
>literature regarding this aspect.:)

I don't think this is what David meant.
Masses only affect kinetic properties.
They do not enter in thermodynamics and therefore
have no influence on the free-energy.

The issue here is if the bonded interaction with the otherwise
dummy masses affect the free energy.
Well, if these bonded interaction do not couple to the rest of
the system not.
In most cases you will have one or more dummy atoms connected
only by one bond to the rest of the molecule.
Then the bond, angle and dihedral interactions do not affect
the ensemble of the rest of the system.
Problems occur when you would for instance eliminate
one atom in a phenyl ring. Then it is clear that the bonded
interactions with the dummy do affect the ensemble of
the rest of the system.

Berk.


>
>Regards
>
>Maik Goette, Dipl. Biol.
>Max Planck Institute for Biophysical Chemistry
>Theoretical & computational biophysics department
>Am Fassberg 11
>37077 Goettingen
>Germany
>Tel.  : ++49 551 201 2310
>Fax   : ++49 551 201 2302
>Email : mgoette[at]mpi-bpc.mpg.de
>         mgoette2[at]gwdg.de
>WWW   : http://www.mpibpc.gwdg.de/groups/grubmueller/
>
>
>David Mobley wrote:
>>Just out of curiousity, since you seem to know something about this, 
>>Maik... If you don't turn off the mass, doesn't it mean that there are 
>>energetic contributions from (for example) bond vibrations and things 
>>which are still left in in the end state? I usually do absolute binding 
>>free energy calculations so I'm not too up on how people correct for the 
>>fact that they have non-interacting dummy atoms which are still bonded to 
>>things in the end state.
>>
>>Thanks,
>>David
>>
>>
>>On 2/6/06, *Maik Goette* <mgoette at mpi-bpc.mpg.de 
>><mailto:mgoette at mpi-bpc.mpg.de>> wrote:
>>
>>     Hi Matt
>>
>>     I am not sure, if there may occur errors because of MW is used as
>>     water-dummy.
>>     But it's really no big deal to insert a DUM with 0 mass and charge.
>>     You can also create as much dummies with the masses you want as you
>>     like.
>>     If you like, I can send you my entries...but as I said, it's really 
>>no
>>     problem to make some.
>>     Nevertheless, you shouldn't make the mass disappear. just turn of VdW
>>     and electorstatics.
>>
>>     Regards
>>
>>     Maik Goette, Dipl. Biol.
>>     Max Planck Institute for Biophysical Chemistry
>>     Theoretical & computational biophysics department
>>     Am Fassberg 11
>>     37077 Goettingen
>>     Germany
>>     Tel.  : ++49 551 201 2310
>>     Fax   : ++49 551 201 2302
>>     Email : mgoette[at]mpi- bpc.mpg.de <http://bpc.mpg.de>
>>              mgoette2[at]gwdg.de
>>     WWW   : http://www.mpibpc.gwdg.de/groups/grubmueller/
>>
>>
>>     mernst at tricity.wsu.edu <mailto:mernst at tricity.wsu.edu> wrote:
>>      > Dear gmx-users,
>>      >
>>      > The Pande group amber port does not contain any specially
>>     designated dummy atom type
>>      > like DUM. It does however contain atom types that seem very
>>     similar, like MW. I have
>>      > used MW as a dummy in a simulation where a hydrogen disappears
>>     (turns into a dummy
>>      > atom), assigning it zero charge as well. Is this a reasonable
>>     substitution? It is my
>>      > understanding that dummy atoms should have zero mass, zero
>>     charge, and zero vdw
>>      > interactions, and that their bonded interactions are unimportant
>>     (can duplicate
>>      > parameters of the atoms they are replacing) since in any case
>>     they will have no
>>      > energetic impact, being massless in their final state. Of course
>>     there are no long-range
>>      > interactions either with dispersion and electrostatic effects set
>>     to zero.
>>      >
>>      > Finally, I would like to thank everyone who has responded to my
>>     questions over the last
>>      > few days. Your help has been invaluable in getting me up to speed
>>     in this new simulation
>>      > environment.
>>      >
>>      > Matt Ernst
>>      > Washington State University
>>      >
>>      > _______________________________________________
>>      > gmx-users mailing list     gmx-users at gromacs.org
>>     <mailto: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
>>     <mailto:gmx-users-request at gromacs.org>.
>>      > Can't post? Read http://www.gromacs.org/mailing_lists/users.php
>>     <http://www.gromacs.org/mailing_lists/users.php>
>>      >
>>      > .
>>      >
>>     _______________________________________________
>>     gmx-users mailing list    gmx-users at gromacs.org
>>     <mailto: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
>>     <mailto: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 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 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