[gmx-users] Extracting enthalpies during or after MD

[Mark Abraham]

----- Original Message -----
From: lists at jsx.dk
Date: Saturday, June 5, 2010 5:58
Subject: Re: [gmx-users] Extracting enthalpies during or after MD
To: Discussion list for GROMACS users <gmx-users at gromacs.org>

> Thanks for the help Mark. Just to make sure I understand you 
> correctly, I
> have a few comments/questions, see below...
> 
> kind regards,
> Jesper
> 
> > ----- Original Message -----
> > From: lists at jsx.dk
> > Date: Friday, June 4, 2010 10:35
> > Subject: Re: [gmx-users] Extracting enthalpies during or after MD
> > To: Discussion list for GROMACS users <gmx-users at gromacs.org>
> >
> >> Hi Mark,
> >>
> >> Thanks for the quick reply.
> >>
> >> The thing with the energy groups makes sense now. But I have 
> run into
> >> another problem with the method...
> >>
> >> The same atom cannot be in multiple energy groups during a 
> simulation.>
> > It doesn't want to be... see manual 3.3
> 
> And there is no way to force it to be, right?

Correct. The nonbonded loops tend to be (roughly) loops over energy groups over charge groups.

> >> Let say that I want the interaction energy between water and lipid
> >> head-group and also water and protein. Will I then have to do 
> "mdrun>> -rerun" to get all the energy group I want out of this?
> >
> > One (re)run will suffice unless you have large numbers of 
> groups. Consider
> > whether g_enemat is useful for you.
> 
> 
> Using rerun will I be able to specify more energy groups, where 
> the same
> atom can be in multiple groups? Else, I would have to run it several
> times, right?

No. Read manual 3.3 carefully - "mutual interactions between all energy monitor groups are compiled during the simulations". So you can get protein-lipid-head-group and protein-solvent (and the other pairs, including self-self, unless you use energy group exclusions) merely by defining three non-intersecting energy groups.

> Also, g_enemat is to post-process/analyse the numbers once I 
> have them
> from a simulation, where I have extracted all the energy numbers 
> I want,
> right?

Yes. g_enemat -h.

> So if I cannot do it with one simulations (because some atoms 
> are part of
> several different energy groups, then g_enemat won't work for my case,
> correct?

You can do it with one invocation of mdrun, which is why I suggested that :-)

Mark

> >> And does mdrun -rerun work on a parallel cluster?
> >
> > Yes, it's exactly like mdrun, but derives the next positions 
> from the
> > trajectory, not from integration. If your groups are small, 
> then use of
> > energy group exclusions will make the problem small enough 
> that you don't
> > need parallelism, however.
> >
> > Obviously these interaction energies won't mean anything if 
> you're using
> > PME, since the long-range term can't be decomposed.
> >
> > Mark
> >
> >> > ----- Original Message -----
> >> > From: lists at jsx.dk
> >> > Date: Friday, June 4, 2010 8:47
> >> > Subject: [gmx-users] Extracting enthalpies during or after MD
> >> > To: gmx-users at gromacs.org
> >> >
> >> >> Hello,
> >> >>
> >> >> I am wondering if it is possible to output interaction
> >> >> energies/enthalpiesduring an MD simulation between specific
> >> >> groups of atoms.
> >> >
> >> > Yes. Look up "energy groups" in the manual.
> >> >
> >> >> The energies are all calculated anyways before the forces are
> >> >> calculated,so I am wondering if there is an option to output
> >> >> those energies to a
> >> >> file?
> >> >
> >> > They'll go to the .edr file and be accessible with (e.g.) 
> g_energy>> >
> >> >> I saw someone on the list who had re-coded gromacs to 
> output these
> >> >> energies, but this resulted in a significant slowdown of the
> >> >> simulationbecause he made the code perform output in an 
> inner-loop.
> >> >
> >> > Yep, don't do that.
> >> >
> >> >> If this is not an option, is there a clever way to get these
> >> >> numbers out
> >> >> after having run the simulation. Essentially, what I want 
> is to
> >> >> get the
> >> >> enthalpy, between e.g. a PO4 head-group bead and a C1A 
> tail-bead
> >> >> using the
> >> >> Martini FF.
> >> >
> >> > You can implement the above to happen during your simulation,
> >> or use the
> >> > facility of mdrun -rerun on a previously computed trajectory
> >> with a
> >> > suitable .tpr to recompute only the energies of the groups of
> >> interest .
> >> >
> >> > Mark
> >> >
> >> > --
> >> > gmx-users mailing list    gmx-users at gromacs.org
> >> > http://lists.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://lists.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://lists.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://lists.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/20100605/b98427ec/attachment.html>