Re: membrane equilibration

From: Axel Kohlmeyer (akohlmey_at_gmail.com)
Date: Thu Mar 03 2011 - 20:42:59 CST

On Thu, Mar 3, 2011 at 7:30 PM, jeremy adler <jeremyeadler_at_gmail.com> wrote:
> Hi Dir. Harrison,
>
> these are the last few lines of minimization (1000 steps with virtually no
> change in the energies. If the equilibration in charmm was carried out over
> about 1500ps behaved itslef perfectly well in the subsequent production
> run.. As potentially another solution, do you know of a script that will
> convert charmm restart files into namd restart files?

jeremy,

you have gigantic energies and forces in your minimization.
no surprise your MD gives in. you should very carefully check,
if you transfer all parameters correctly from your charmm simulation
to your namd input. you already suspected the box dimensions.
those are often but not always a cause of these sysmptoms.

why don't you load the psf and restart into vmd and check
where atoms 3241 and 5379 are located (remember to provide
cell information via "pbc set" and turn on periodic display)
and see if there are any detectable overlaps.

cheers,
   axel.

> thanks!
>
> Jeremy
>
> starting system:
> http://www.mediafire.com/file/iq1q1z96ac79nje/step6.6_equilibration.pdb
>
> BRACKET: 2.85511e-09 31.25 -9.12482e+11 -9.07398e+11 -8.99207e+11
> PRESSURE: 998 3.88841e+16 -2.14232e+16 1.12373e+17 3.04533e+16 -2.00369e+16
> 2.26124e+16 7.48332e+15 -6.20386e+15 1.91964e+16
> GPRESSURE: 998 3.88841e+16 -2.14232e+16 1.12373e+17 3.04533e+16 -2.00369e+16
> 2.26124e+16 7.48332e+15 -6.20386e+15 1.91964e+16
> ENERGY: 998 2346753.6788 468115.1150 4454.7007
> 4848.7842 -31018.7247 99999999.9999 0.0000
> 0.0000 0.0000 99999999.9999 0.0000 99999999.9999
> 99999999.9999 0.0000 99999999.9999 99999999.9999
> 23040.0000 99999999.9999 99999999.9999
>
> BRACKET: 1.76455e-09 16.1875 -9.12482e+11 -9.07398e+11 -9.04265e+11
> PRESSURE: 999 3.88841e+16 -2.14232e+16 1.12373e+17 3.04533e+16 -2.00369e+16
> 2.26124e+16 7.48332e+15 -6.20386e+15 1.91964e+16
> GPRESSURE: 999 3.88841e+16 -2.14232e+16 1.12373e+17 3.04533e+16 -2.00369e+16
> 2.26124e+16 7.48332e+15 -6.20386e+15 1.91964e+16
> ENERGY: 999 2346695.8646 468093.2401 4454.6479
> 4848.7877 -31019.1667 99999999.9999 0.0000
> 0.0000 0.0000 99999999.9999 0.0000 99999999.9999
> 99999999.9999 0.0000 99999999.9999 99999999.9999
> 23040.0000 99999999.9999 99999999.9999
>
> BRACKET: 1.09055e-09 3.0625 -9.09338e+11 -9.07398e+11 -9.04265e+11
> NEW SEARCH DIRECTION
> INITIAL STEP: 2.5e-07
> GRADIENT TOLERANCE: 1.89282e+08
> TIMING: 1000 CPU: 2751.05, 2.03812/step Wall: 2751.05, 2.03812/step, 0
> hours remaining, 65648 kB of memory in use.
> PRESSURE: 1000 3.88841e+16 -2.14232e+16 1.12373e+17 3.04533e+16 -2.00369e+16
> 2.26124e+16 7.48332e+15 -6.20386e+15 1.91964e+16
> GPRESSURE: 1000 3.88841e+16 -2.14232e+16 1.12373e+17 3.04533e+16
> -2.00369e+16 2.26124e+16 7.48332e+15 -6.20386e+15 1.91964e+16
> ETITLE: TS BOND ANGLE DIHED
> IMPRP ELECT VDW BOUNDARY
> MISC KINETIC TOTAL TEMP
> TOTAL2 TOTAL3 TEMPAVG PRESSURE
> GPRESSURE VOLUME PRESSAVG GPRESSAVG
>
> ENERGY: 1000 2346695.8646 468093.2401 4454.6479
> 4848.7877 -31019.1667 99999999.9999 0.0000
> 0.0000 0.0000 99999999.9999 0.0000 99999999.9999
> 99999999.9999 0.0000 99999999.9999 99999999.9999
> 23040.0000 99999999.9999 99999999.9999
>
> WRITING EXTENDED SYSTEM TO RESTART FILE AT STEP 1000
> OPENING COORDINATE DCD FILE
> WRITING COORDINATES TO DCD FILE AT STEP 1000
> WRITING COORDINATES TO RESTART FILE AT STEP 1000
> FINISHED WRITING RESTART COORDINATES
> WRITING VELOCITIES TO RESTART FILE AT STEP 1000
> FINISHED WRITING RESTART VELOCITIES
> REINITIALIZING VELOCITIES AT STEP 1000 TO 300 KELVIN.
> TCL: Running for 250000 steps
> ERROR: Constraint failure in RATTLE algorithm for atom 3241!
> ERROR: Constraint failure; simulation has become unstable.
> ERROR: Constraint failure in RATTLE algorithm for atom 5379!
> ERROR: Constraint failure; simulation has become unstable.
> ERROR: Exiting prematurely.
> ==========================================
> WallClock: 2760.421875 CPUTime: 2760.421875 Memory: 65648 kB
> End of program
>
> On Thu, Mar 3, 2011 at 7:10 PM, Chris Harrison <charris5_at_gmail.com> wrote:
>>
>> Were you using multiple-time-stepping in CHARMM? I suspect not, and
>> that you are in NAMD. If this is the case, you could have some
>> mildly pathological interactions that are not updated as often
>> and lead to strong velocities that result in RATTLE failures. Some
>> solutions are to either determine the specific atom (via it's index) and
>> resolve whatever may be causing the instability, or you could attempt
>> longer minimization/equilibration runs in hopes the unstable interaction
>> resolves itself. If this is indeed your situation, you'll be more
>> efficient if you re-run the crashing simulation, but outputting the
>> frames and energies every step, and then analyzing that trajectory for
>> the source of your RATTLE failure.
>>
>> HOWEVER, all the above is just a best guess based on your mention of a
>> RATTLE failure. Axel is right that it's difficult to diagnose the problem
>> without seeing the actual output in your query. It's often useful to
>> cut and paste 20 or so lines around the error message, so others can see
>> exactly the output. With that kind of data the namd-list can be
>> much more helpful.
>>
>> Best,
>> Chris
>>
>>
>> --
>> Chris Harrison, Ph.D.
>> Theoretical and Computational Biophysics Group
>> NIH Resource for Macromolecular Modeling and Bioinformatics
>> Beckman Institute for Advanced Science and Technology
>> University of Illinois, 405 N. Mathews Ave., Urbana, IL 61801
>>
>> char_at_ks.uiuc.edu Voice: 217-244-1733
>> http://www.ks.uiuc.edu/~char Fax: 217-244-6078
>>
>>
>> jeremy adler <jeremyeadler_at_gmail.com> writes:
>> > Date: Thu, 3 Mar 2011 18:03:18 -0500
>> > From: jeremy adler <jeremyeadler_at_gmail.com>
>> > To: Axel Kohlmeyer <akohlmey_at_gmail.com>
>> > Cc: namd-l_at_ks.uiuc.edu
>> > Subject: Re: namd-l: membrane equilibration
>> >
>> > Hi Dr Kohlmeyer,
>> >
>> > It seems as though the energies never come down and I get "ERROR:
>> > Constraint
>> > failure in RATTLE algorithm" when the program exits the minimization
>> > portion. Ive tried using both fixed atom as well as harmonic restraints
>> > with
>> > the same basic result. I had seen somewhere in the mailing list that
>> > periodic boundary conditions might be too small so i made them bigger
>> > only
>> > to have the water collapse into the membrane. Im not exactly clear why
>> > the
>> > energies are so he given they werent nearly as high when i was using
>> > charmm
>> > and it equilibrated fine then (i am using the preequilbrated membrane
>> > from a
>> > charmm run i conducted). I would like to use namd because i have access
>> > to a
>> > bluegene computer and namd functions much better on bluegene than
>> > charmm.
>> >
>> >
>> > thanks for your help,
>> >
>> > Jeremy
>> >
>> > On Thu, Mar 3, 2011 at 2:43 PM, Axel Kohlmeyer <akohlmey_at_gmail.com>
>> > wrote:
>> >
>> > > On Thu, Mar 3, 2011 at 2:02 PM, jeremy adler <jeremyeadler_at_gmail.com>
>> > > wrote:
>> > > > Hi
>> > > >
>> > > > I was wondering if anyone has a membrane equilbration protocol that
>> > > works. I
>> > > > tried adapting the one in the tutorial with a membrane previousely
>> > > > equilibrated in charmm to no avail
>> > >
>> > > jeremy,
>> > >
>> > > before stating that something doesn't work, you should explain
>> > > how it fails for you and why you think that this is not adequate.
>> > >
>> > > in molecular simulations, there are so many little things that
>> > > can matter, it is rare that there is a one-size-fits-all solution.
>> > >
>> > > cheers,
>> > > axel.
>> > >
>> > > >
>> > > > thanks
>> > > >
>> > > > Jeremy
>> > > >
>> > >
>> > >
>> > >
>> > > --
>> > > Dr. Axel Kohlmeyer
>> > > akohlmey_at_gmail.com http://goo.gl/1wk0
>> > >
>> > > Institute for Computational Molecular Science
>> > > Temple University, Philadelphia PA, USA.
>> > >
>
>

-- 
Dr. Axel Kohlmeyer
akohlmey_at_gmail.com http://goo.gl/1wk0
Institute for Computational Molecular Science
Temple University, Philadelphia PA, USA.

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:56:42 CST