[gmx-users] Problem in energy minimization and further dynamics probably for atomic clashes

Thu Jun 19 09:58:33 CEST 2014

The size of a step you can take is intrinsically tied to the typical size
of the forces being integrated. We equilibrate to get the coordinates into
a region of phase space that is meaningful, and this typically has the side
effect of producing forces that are smaller and so can be integrated over
larger time steps. This warning is intended to prevent a class of silly
errors related to this phenomenon. If the arbitrary force constant on an
interaction is too large then you have to take smaller steps or produce
junk integration. Or choose differently, perhaps based on your literature
reading about simulating such systems ;-)

Mark

On Thu, Jun 19, 2014 at 7:44 AM, sucharita dey <sucharita.dey at gmail.com>
wrote:

> ok, I tried with an arbitrary value of kb for all the harmonic bonds
> (7150000) and the distances as they are in the crystal structure.  There
> are 5 harmonic bonds with FE which I defined in the itp, but I am getting
> an error on one of them while doing position restrained MD for 50 ps. I
> tried with both the constraints = none / hbond , but getting the same
> error. (Here, I have turned on the temperature coupling as well)
>
>
>
>
> *The bond in molecule-type Protein_chain_A between atoms 4207 FE and 2546
> NE2 has an estimated oscillational period of 7.9e-03 ps, which is less
> than 5 times the time step of 2.0e-03 ps.  Maybe you forgot to change the
> constraints mdp option.*
>
> Can you please suggest where might be the problem?
>
> Thanks,
> Sucharita
>
>
> On Wed, Jun 18, 2014 at 7:54 PM, Justin Lemkul <jalemkul at vt.edu> wrote:
>
> >
> >
> > On 6/18/14, 6:47 AM, sucharita dey wrote:
> >
> >> Thanks Justin. The OGA topology is now OK. But still the problem with
> the
> >> FE ion persists (very high force on it) during the minimization.
> >> Yes, I want to link the FE with its co-ordination atoms by simple
> harmonic
> >> (bond type 6), can you please tell what will be the value for kb (the
> >> force
> >> constant)?
> >>
> >>
> > No idea.  If there are no vibrational data available for the interaction,
> > you're stuck with using some arbitrary value.  That's actually somewhat
> > common when dealing with transition metals, because that's the least of
> > your worries when it comes to such species; MM representations of such
> > metals are very poor in general.
> >
> > -Justin
> >
> >
> >
> >> On Fri, Jun 13, 2014 at 8:41 PM, Justin Lemkul <jalemkul at vt.edu> wrote:
> >>
> >>
> >>>
> >>> On 6/13/14, 2:11 AM, sucharita dey wrote:
> >>>
> >>>  Dear Users,
> >>>>
> >>>> I have a protein-DNA system with 3 Zinc fingers, 1 Fe++, 1 cofactor
> NOG
> >>>> (N-OxalyGlycine) and crystal waters.
> >>>>
> >>>> I have generated the initial topology of OGA from PRODRG and
> >>>> incorporated
> >>>> it in the  the forcefield gromos 53a6 and since the forcefield already
> >>>> has
> >>>> parametrs for ZN and FE, I had no problem in generating the  *.itp
> >>>> files.
> >>>> I
> >>>> solvated the system and ran steep minimization for 1000 steps with
> >>>> emstep
> >>>> =
> >>>> 0.1 and emtol 1.
> >>>>
> >>>>
> >>>>  Is the NOG topology sound?  PRODRG has known problems getting charges
> >>> right, but building a suitable NOG topology from existing building
> blocks
> >>> is trivial.
> >>>
> >>>
> >>>   It ran for ~80 steps and stopped. I checked the gro file generated,
> the
> >>>
> >>>> problem is around OGA and the FE,  and the OGA is loosing its
> >>>> structure. I
> >>>> suspect it due to clashes, -- actually in the crystal structure 2
> >>>> oxygens
> >>>> of OGA are in co-ordination with the FE (the FE being co-ordinated
> with
> >>>> 4
> >>>> other atoms including one O from crystal water).
> >>>>
> >>>>
> >>>>  Sounds like a potential topology issue.  One simple test is to run
> NOG
> >>> in
> >>> vacuo then in a box of water to see if it is stable on its own, then
> deal
> >>> with it in the context of the full protein+ions.
> >>>
> >>>
> >>>   I have not considered the FE co-ordination, can you please suggest
> how
> >>> to
> >>>
> >>>> do this or else please suggest if you feel the problem is elsewhere.
> >>>>
> >>>>
> >>>>  Ion coordination is easily done with distance restraints or type-6
> >>> harmonic connections.
> >>>
> >>>   Below is the comment given after minimization stopped:
> >>>
> >>>>
> >>>> *Energy minimization has stopped, but the forces havenot converged to
> >>>> therequested precision Fmax < 1 (whichmay not be possible for your
> >>>> system).
> >>>> Itstoppedbecause the algorithm tried to make a new step whose sizewas
> >>>> toosmall, or there was no change in the energy sincelast step. Either
> >>>> way,
> >>>> weregard the minimization asconverged to within the available
> >>>> machineprecision,given your starting configuration and EM
> >>>> parameters.Double
> >>>>
> >>>> precision normally gives you higher accuracy, butthis is often
> notneeded
> >>>> for preparing to run moleculardynamics.writing lowest energy
> >>>> coordinates.Steepest Descents converged to machine precision in 84
> >>>> steps,but did not reach the requested Fmax < 1.Potential Energy  =
> >>>> -3.7139549e+08Maximum force     =  7.7602650e+14 on atom 4207Norm of
> >>>> force     =  3.6179670e+12*
> >>>>
> >>>> NOTE: the atom 4207 with maximum force is FE
> >>>>
> >>>>
> >>>>  You have essentially infinite forces, which indicate very bad
> geometry,
> >>> clashes, or a bad topology.
> >>>
> >>> -Justin
> >>>
> >>> --
> >>> ==================================================
> >>>
> >>> Justin A. Lemkul, Ph.D.
> >>> Ruth L. Kirschstein NRSA Postdoctoral Fellow
> >>>
> >>> Department of Pharmaceutical Sciences
> >>> School of Pharmacy
> >>> Health Sciences Facility II, Room 601
> >>> University of Maryland, Baltimore
> >>> 20 Penn St.
> >>> Baltimore, MD 21201
> >>>
> >>> jalemkul at outerbanks.umaryland.edu | (410) 706-7441
> >>> http://mackerell.umaryland.edu/~jalemkul
> >>>
> >>> ==================================================
> >>> --
> >>> 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.
> >>>
> >>>
> > --
> > ==================================================
> >
> > Justin A. Lemkul, Ph.D.
> > Ruth L. Kirschstein NRSA Postdoctoral Fellow
> >
> > Department of Pharmaceutical Sciences
> > School of Pharmacy
> > Health Sciences Facility II, Room 601
> > University of Maryland, Baltimore
> > 20 Penn St.
> > Baltimore, MD 21201
> >
> > jalemkul at outerbanks.umaryland.edu | (410) 706-7441
> > http://mackerell.umaryland.edu/~jalemkul
> >
> > ==================================================
> > --
> > 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.
>