Re: 2.7b metadynamics ::alpha module :: folding of a single helix in water

From: Giacomo Fiorin (gfiorin_at_seas.upenn.edu)
Date: Wed Jun 10 2009 - 08:57:38 CDT

Hi Ali, I see: what you mean (narrowing a1 to an interval of interest) is a
good idea, but to do so you also need to set some force constants at
the two boundaries. Otherwise, those are just the limits of the grid, and
they get expanded over time as the colvar moves outside of them. The
fact that lowerBoundary goes negative can be worrisome, but it is only
due to the
fact that the code managing the grid doesn't know that a1 can't go
negative, and allocates a buffer region.

Of course, it's never 100% guaranteed that the code is free of bugs,
especially at this stage. So, if you do observe negative values of
the alpha variable in the .colvars.traj file, I'd like to know.

You're applying small hills (0.04 kcal/mol) but very frequently (10
steps), so you *are* pumping energy into it. Maybe the helix is just
too stiff and by the time you've added energy to one structure
metadynamics has moved it back to a similar one?

This is why I suggest that you try the harmonic steering: you could,
say, set a target of 1 to try and fold the helix completely. Unlike
in metadynamics, at least you'd drive it monotonically there, and
you'd have to cope with less complications.

Otherwise, if your target is the folded helix, and the other
structures are not relevant to you, it may simply be happening that
because of the system's nature the colvar does not respond as quickly
as you want it to, because other degrees of freedom need to be forced
(i.e. additional colvars).

Ciao
Giacomo

---- ----
 Giacomo Fiorin
   Center for Molecular Modeling at
     University of Pennsylvania
     231 S 34th Street, Philadelphia, PA 19104-6323
 phone: (+1)-215-573-4773
 fax: (+1)-215-573-6233
 mobile: (+1)-267-324-7676
 mail: giacomo.fiorin_<at>_gmail.com
 web: http://www.cmm.upenn.edu/
---- ----

On Wed, Jun 10, 2009 at 1:02 AM, Ali Emileh<ali.emileh_at_gmail.com> wrote:
> Hi,
>
> About the warning messages :
> My concern is that it seems the amount of time/force I'm giving the system
> to 'diffuse' along
> the colvar is not enough. So I was thinking by limiting it to build up the
> Gaussians in a
> restricted region, I'll get the system to change (saturate the energy wells)
> in a shorter
> amount of time. Please correct me if I'm wrong.
>
> This is the complete config file :
> ===================================
> colvarsTrajFrequency 1000
>
> colvar {
> name a1
> width 0.05
> outputVelocity on
> outputAppliedForce on
> lowerBoundary 0.3
> upperBoundary 1.0
>
> alpha {
> residueRange 100-116
> psfSegID G
> }
> }
>
> metadynamics {
> colvars a1
> hillWeight 0.04
> dumpFreeEnergyFile yes
> newHillFrequency 10
> }
> ===================================
> This config is used in a run I have going now. The .traj file doesn't show
> much
> difference after about 0.5 ns.
>
> The peptide is almost half-folded (as the starting value of the colvar
> shows; ~0.4)
> and my guess is the alpha colvar varies 0.0-1.0. Now I don't get the
> negative value
> for the colvar, especially for a partially folded helix. The dox say this
> component
> returns positive values. Is it possible there's a small error in how the
> formula is
> implemented?
>
> I'm not using the extendedLagrangian option and the documentation says it
> should
> be off by default. So I'm assuming thats the case.
>
> About the harmonic restraint, I'll give it a try as soon as I sit down and
> read about
> what it does. Thank you for the helpful work around.
>
> Thank you,
> Ali
>
>
>
> On Tue, Jun 9, 2009 at 1:24 PM, Giacomo Fiorin <gfiorin_at_seas.upenn.edu>
> wrote:
>>
>> Hi Ali, the messages you've found are generated by two minor flaws in
>> the code, which do not affect the dynamics. The Warning about the
>> commensurate boundaries arises improperly, due to how we test that
>> condition in the code, and will be fixed in the next release.
>> Instead, why the boundaries change is because the code tries to
>> accommodate all values that are explored within the run, according to
>> expandGrids. By default this flag is on, which is inconsistent with
>> what I wrote in the documentation: again, this is not your fault. But
>> this flag only changes the length of the PMF anyway (and not its
>> resolution), and most of all does not affect the trajectory that
>> seriously.
>>
>> Can you send the complete configuration?
>>
>> One thing your msg says is that you are indeed exploring low values
>> (little higher than -0.38) of the a1 colvar, well below the 0.3
>> boundary you initially set: the peptide structure should be mostly
>> unfolded near zero.
>>
>> Are you maybe using the fictitious particle of the extended
>> Lagrangian? In that case, what you see moving would not be the
>> physical value, but that of the fictitious particle, absorbing all the
>> energy but not transmitting it to the atoms as it should. The
>> optional extendedLagrangian flag can be useful in some cases, but it's
>> not required to obtain a PMF, and it's often a good idea to ignore it,
>> because things like this could occur if the force constant and mass
>> hit one of the resonance frequencies of the system.
>>
>> If you just want to test the variable, I suggest a simpler approach:
>> skip metadynamics, and do a steering through the harmonic restraint,
>> setting the initial restraint center at the initial value and a target
>> center away from it. You can then control in how many steps to move
>> it there. If you see the trajectory following a straight line with
>> some oscillations, then the colvar works for your system.
>>
>> Bests
>> Giacomo
>>
>> ---- ----
>> Giacomo Fiorin
>> Center for Molecular Modeling at
>> University of Pennsylvania
>> 231 S 34th Street, Philadelphia, PA 19104-6323
>> phone: (+1)-215-573-4773
>> fax: (+1)-215-573-6233
>> mobile: (+1)-267-324-7676
>> mail: giacomo.fiorin_<at>_gmail.com
>> web: http://www.cmm.upenn.edu/
>> ---- ----
>>
>
>

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:51:04 CST