[gmx-users] Re: Re: Alchemistry free energy simulations of molecules with different atom numbers

[Zhao Lifeng]

>
> On Wed, Nov 19, 2008 at 2:14 PM, Zhao Lifeng <zhao.lf at gmail.com> wrote:
>
> >
> > Dear all,
> >
> > I encountered a problem in alchemistry free energy calculations:
> >
> > In my simulations a group (A) is to be changed to (B) in a molecule,
> while
> > the atom numbers n(A) is not equal to n(B).
> > A scheme is to include both A and B in the same topology file and switch
> > on/off the B/A charges and vdw interactions.
> > And interactions between A and B were excluded in the simulations.
> However,
> > the vibration mode would not be correct
> > for both A and B state.
> >
> > Another scheme is to change the corresponding atoms to zero mass dummy
> > atoms or from them. In this case, there will be
> > no interactions on the dummy atoms, either bonded or non-bonded
> > interactions.
> >
> > I don't know if these schemes are correct and if there is any other more
> > reasonable schemes. In my simulations I prefered the
> > later one and thus the problems came.
> > In state of lambda=0 or =1, since there are dummy atoms with no force on
> > them, (sometimes with velocities, e.g. from
> > the last lambda simulation),  the  dummy atoms often bring forth errors,
> > for example, 1-4 distance too large, neighbor list error
> > and segmentation fault. Bond constraints seems work but it might not be
> > reasonable for heavy atoms.
> >
> > Any suggestions would be appreciated.
>
>
> Hello,
>
> I have used schemes similar to the one you described and to circumvent the
> problems associated with growing or disappearing particles out  of nothing
> I
> used soft-core interactions and manipulated the soft-core parameters to get
> "nice" dgdl curves. Hope  this helps.
>
> Diana Lousa
> PhD student
> Protein Modelling  Laboratory
> ITQB/UNL
> Oeiras, Portugal
>
> >
> > Lifeng Zhao
>


Thanks Diana. But softcore does not help for it only affect vdw
interactions.
In my TI simulation case, the dummy atoms has no mass and the bond
interactions
are turned off. (If bond interactions were not turned off there would
produce
a segmentation fault for forces on zero mass.)

Lifeng Zhao
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20081120/a3aac0fa/attachment.html>