[gmx-users] Dummy atoms flopping everywhere

[Raphaël Robidas]

Hello,

Thank you for your suggestion! I tried it and the benzyl isn't flopping
around uncontrollably, and I get much more reasonable FEP energies.

I am still wondering one thing: I used LigParGen to generate the hybrid
topology of my two ligands. I can see that it is morphing the atoms in the
benzyloxy group  to dummy atoms with mass 1.0080 (even carbon atoms):

[ atoms ]
;   nr       type  resnr    res   atom   cgnr     charge       mass
 typeB    chargeB      massB comments
     1   opls_800      1    A2B    H00      1     0.1293     1.0080
 opls_9855     0.0000     1.0080
     2   opls_801      1    A2B    C01      1    -0.1350    12.0110
 opls_9856     0.0000     1.0080
     3   opls_802      1    A2B    C02      1    -0.1431    12.0110
 opls_9857     0.0000     1.0080

...


Is this another technique of dealing with dummy atoms? All the relative
free energy perturbation tutorials I have seen use very small perturbations
(ex. substituants on aryls), so I am not sure exactly how to deal with this
larger change. Any help/pointers are greatly appreciated!

Best

Le mer. 27 nov. 2019 à 03:26, Alessandra Villa <
alessandra.villa.biosim at gmail.com> a écrit :

> Hi,
>
> On Mon, Nov 25, 2019 at 6:20 PM Raphaël Robidas <raphaelrobidas at gmail.com>
> wrote:
>
> > I am doing a free energy perturbation calculation on a protein-ligand
> > system. I am going from a topology A which has a benzyloxy group to a
> > topology B which only has a methoxy group. My problem is that when
> lambda =
> > 1, the simulation crashes because the benzyloxy group is flopping around
> > too fast, since it has no mass nor collision. I am wondering that the
> best
> > way to deal with this would be.
> >
> >
> If  you switch off only the non-bonded interactions when going from A to B,
> this should not happen since  the bond interactions (e.i. torsion) are
> still active for the benzyloxy group (thus the benzyloxy group can not
> freely rotate).
> Best regards
> Alessandra
>
>
> > Using restraints would be an option, but I don't need to restrain my
> ligand
> > otherwise, so it would be desirable to avoid restraints overall. Could a
> > second topology without the dummy atoms be used just when lambda = 1 or
> it
> > won't be compatible with the rest of the simulations?
> >
> > Thank you!
> > --
> > Gromacs Users mailing list
> >
> > * Please search the archive at
> > http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before
> > posting!
> >
> > * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
> >
> > * For (un)subscribe requests visit
> > https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or
> > send a mail to gmx-users-request at gromacs.org.
> >
> --
> Gromacs Users mailing list
>
> * Please search the archive at
> http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before
> posting!
>
> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>
> * For (un)subscribe requests visit
> https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or
> send a mail to gmx-users-request at gromacs.org.