Dear Fabian,
If you are indeed considering absolute free energy calculation, for the
reasons that Hannes advanced, then have a look at EMIL.
starting point:
http://ambermd.org/tutorials/advanced/tutorial19/
This is an extension for TI in sander/pmemd intended to avoid the need to
perturb all the way to nothing, allowing the user to instead perturb merely
to a simplified version of the original Hamiltonian which is analytically
tractable.
Josh
On 19 November 2015 at 17:54, Hannes Loeffler <Hannes.Loeffler.stfc.ac.uk>
wrote:
> Dear Fabian,
>
> what you could try is "absolute" free energy simulation where you
> completely "destroy" the ligand. This means you mutate it into nothing
> and does get the solvation free energy. Have a look
> into http://pubs.acs.org/doi/abs/10.1021/ct400340s how to do that. The
> SI of that paper describes how to prepare inputs with sample scripts.
>
> Your ligands are very dissimilar anyway so computing relative free
> energies by mutating them into each other may be problematic and
> difficult to do. But you ligands are also very large so you should
> carefully monitor if your free energy converges and also look into the
> statistics.
>
> Cheers,
> Hannes.
>
>
> On Thu, 19 Nov 2015 18:40:19 +0200
> Fabian gmail <fabian.glaser.gmail.com> wrote:
>
> > Dear Hannes,
> >
> > Sorry to bother you, I went through the A9 tutorial in detail, and I
> > have several additional questions to my specific goals and molecules:
> >
> > I am not sure how to construct my thermodynamic cycle, in my case
> > there is no protein, only two different ligands, for what I saw in
> > your tutorial cycle, you need a protein medium to run MD and mutate
> > the ligand. In my case the protein medium does not exist and the
> > ligands are very big and quite different… I would prefer not to
> > mutate them if possible.
> >
> > Another possibility for me would be to run two different type of
> > solvents, one would be the ligand in pure water and the other in
> > water + acetone for example and calculate the difference between the
> > two ligands. And then I would calculate the DG of transfer between
> > water to water + acetone or the hydration in water + acetone, which
> > is very close to what really happens, Is this doable with TI?
> >
> > Additionally, the ligands I consider are very very flexible large and
> > different see them:
> >
> > https://en.wikipedia.org/wiki/Darunavir
> > <https://en.wikipedia.org/wiki/Darunavir>
> > https://en.wikipedia.org/wiki/Ritonavir
> > <https://en.wikipedia.org/wiki/Ritonavir>
> >
> > Can they be mutated into each other without accumulating charge
> > errors?
> >
> > I think I would prefer not to mutate them but to calculate the
> > hydration free energies of each of them in water and / or in water +
> > acetone mixtures.
> >
> > Is that a good approach?
> >
> > Thansk a lot again,
> >
> > Fabian
> >
> >
> > Dr. Fabian Glaser
> > Head of the Structural Bioinformatics section
> >
> > Bioinformatics Knowledge Unit - BKU
> > The Lorry I. Lokey Interdisciplinary Center for Life Sciences and
> > Engineering Technion - Israel Institute of Technology, Haifa 32000,
> > ISRAEL
> >
> > fglaser at technion dot ac dot il
> > Tel: +972 4 8293701
> > http://bku.technion.ac.il
> >
> >
> > > On 17 Nov 2015, at 6:06 PM, Hannes Loeffler
> > > <Hannes.Loeffler.stfc.ac.uk> wrote:
> > >
> > > On Tue, 17 Nov 2015 17:35:04 +0200
> > > Fabian gmail <fabian.glaser.gmail.com> wrote:
> > >
> > >> Dear Hannes,
> > >>
> > >> Thanks a lot, it looks very interesting and relevant, my molecules
> > >> are much larger (ritonavir and darunavir) but still I maybe able to
> > >> use similar ideas.
> > >
> > > Regarding the number of atoms the following may be of use
> > >
> https://mailman-1.sys.kth.se/pipermail/gromacs.org_gmx-users/2015-July/099194.html
> > >
> > > _______________________________________________
> > > AMBER mailing list
> > > AMBER.ambermd.org
> > > http://lists.ambermd.org/mailman/listinfo/amber
> >
> > _______________________________________________
> > AMBER mailing list
> > AMBER.ambermd.org
> > http://lists.ambermd.org/mailman/listinfo/amber
>
>
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Sun Nov 22 2015 - 23:30:03 PST
