[gmx-users] Re: question about free energy. Gromacs user

[David Mobley]

Hi,

> I just have a couple of things i want to clarify about the constrain
> distances i used.  The distance I used to pull the ligand and receptor
> apart was a non-bonded distance between the amide nitrogen of the ligand
> and the oxygen carbonyl of the receptor (on both ends).  looking at the
> slowgrowth results of my 5 ns, the distances i specifed increased from
> lambda 0 to 1, however, I didnt get separation as I wanted it since the
> ends just separated and the middle interaction of the ligand-receptor
> stayed intact. nothing unusal happened to the ligand or receptor.
> I dont think my method altered the internal degrees of freedom of the
> ligand. Can you please clarify this issue for me.

In terms of altering the internal degrees of freedom of something, let
me just explain what can go wrong, and you can figure out if it
applies to what you're doing. Basically, suppose you have a ligand
which prefers to be in one conformation when it's bound, and in a
different conformation when it's unbound. Suppose then you apply
constraints, based on the bound structure, to constrain the distance
between atom A in the ligand and reference atom C in the protein, and
to constrain the distance between atom B in the ligand and reference
atom C in the protein. Suppose further that atoms A and B want to be
5A apart when the ligand is in the bound conformation, but 8A when the
ligand is in the unbound conformation. Naturally, looking at the bound
structure, you set your constraints A-C and B-C to pull at the same
rate. Now, the problem occurs if the constraints you've introduced to
the A-C and B-C distances make it hard for the A-B distance to change,
so that when the ligand is pulled out of the protein, distance A-B
remains 5A, though the ligand would prefer to have it be 8A.

The question you have to answer is whether there is any potential that
your constraints would cause this to happen. My hunch is yes: Although
in principle if you constrain A-C and B-C, A-B is not constrained,  in
practice, constraining A-C and B-C probably introduces additional
barriers into the ligand conformational change landscape. I am not
sure though. This is what you have to figure out.

If the constraints do affect the A-B distance in the sense I have
described, then your results will probably be garbage.

David


> I used constraint=all-bonds in my mdp file.
>
> Again, thanks you both for the comments.
>
> Belquis
>
>
>
> >
> >>
> >> Maybe it should be obvious, but (a) why are you constraining two
> >> distances, and (b) are you sure your constraints aren't going to muck
> >> with the internal degrees of freedom for the ligand? I would think one
> >> would like to pull the ligand out of the receptor along some
> >> particular direction, but without doing anything to alter its internal
> >> degrees of freedom, or there would be a free energy associated with
> >> changing its internal degrees of freedom that you wouldn't capture in
> >> the PMF calculation.
> >>
> >
> > I agree with David's comments here.  Check out the 'Special Topic'
> > section
> > of the GMX manual.   you can either do a pulling simulation or umbrella
> > sampling (where the position and constrain the ligand on some rxn
> > coordinate, in several simulation).  The output of such simulatsions is a
> > *.pdo file, which you can then run g_wham on to obtain a PMF.
> > _______________________________________________
> > 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
>