Re: [AMBER] constant pressure

From: Yong Duan <duan.ucdavis.edu>
Date: Sun, 04 Dec 2011 12:43:32 -0800

Looks like you did following, correct me if I am wrong,

1) minimization by 100 steps (recorded in min2.dat), final RMS = 3.96
kcal/mole. Unconstrained, ntb=1.
2) From above minimized structure, you started 1000-step MD, as in
heat1.dat. dt=0.5 fs, ntb=1, ntc=1, residues 1-44 constrained, temp0=50.0.
You ran this using 8 CPUs.
3) You then re-started the simulation from 2) for 1000 steps, again,
dt=0.5 fs, ntb=1, ntc=1, residues 1-85 constrained, temp0=50.0.
4) You then re-started the simulation from 3). You switched to ntb=2,
dt=1fs. Here, you ran into trouble because the energy changed a lot. At
the end of 3), VDW=4663 and changed to 132571, indicating you had a couple
of atoms too close to each other.

It appears that you had ntb=1 or ntb=2 for all of the calculations. So,
the PBC appears to be ok.

Can you run step 3) for 100 ps? Just run it longer, much longer, with
exactly the same input files. After that, show us the last line of the
restart file. We go from there. 

--
Yong Duan, Ph.D, Professor
UC Davis Genome Center and
Department of Biomedical Engineering
University of California at Davis
Davis, CA 95616
530-754-7632
On 12/4/11 11:52 AM, "kirtana S" <skirtana4.gmail.com> wrote:
>When I used low temperature simulation , I have attached the output files
>of the minimization and equilibration (I used minimization of solvent
>followed by minimizationof entire system , equilibration at 50K at
>constant
>volume and then I turn on constant pressure equilibration) I also tried
>skipping my last minimization step still I get difference in the
>energetics
>and error message .
>
>As said "Did you turn on the non-bonded interactions of the ion when you
>minimized the structure (before you started the MD simulation)? " , do you
>refer to any input file parameter here. Can you tell me which parameter I
>should use .I have attached the files if you have any suggestion, I
>appreciate your help.
>
>Thanks
>Kirtana
>On Sun, Dec 4, 2011 at 2:14 PM, Brian Radak <radak004.umn.edu> wrote:
>
>> This is probably unlikely, but is there a bad overlap near the periodic
>> boundary?  Minimization usually (always?) ignores PBCs.  If a molecule
>>is
>> immediately being wrapped through the wall onto another molecule then
>> things won't be happy.  That could maybe be happening during the second
>> minimization after the NVT dynamics, which you probably shouldn't need.
>>
>> Brian
>>
>> On Sun, Dec 4, 2011 at 1:52 PM, Yong Duan <duan.ucdavis.edu> wrote:
>>
>> >
>> > There are many reasons that things like what you observed could
>>happen.
>> If
>> > you provide us a bit more detail, we might be able to find out why and
>> > make suggestions.
>> >
>> >
>> >
>> > Did you turn on the non-bonded interactions of the ion when you
>>minimized
>> > the structure (before you started the MD simulation)? The differences
>>are
>> > not limited to non-bonded energy terms. Instead, all energy terms in
>> > minimization and in MD were really different. For example, bond energy
>> > more than doubled from 6004.1653 in minimization to 12704.4067 in the
>> > first step of MD (as indicated by your TIME(PS) = 0.000). Usually,
>>when
>> > you start a simulation from a structure that was minimized, the
>>energies
>> > should be reasonably close at the first step. In fact, we would be
>> > concerned if they are not. If you hadn't told me that they are the
>>same
>> > system, I would have thought they are completely different systems.
>> >
>> > How low was your "low" temperature? Your output says it is 244K. Was
>>this
>> > close to the your "low" temperature? This temperature is considered
>>low
>> in
>> > conventional sense, below freezing point of water. However, in terms
>>of
>> > kinetic energy per atom, 244K is really not much lower than the 300K
>> (only
>> > about 80%). It sometimes helps to start simulations from 10K,
>>although,
>> in
>> > this case, I am not sure this would help you much.
>> >
>> >
>> > --
>> > Yong Duan, Ph.D, Professor
>> > UC Davis Genome Center and
>> > Department of Biomedical Engineering
>> > University of California at Davis
>> > Davis, CA 95616
>> > 530-754-7632
>> >
>> >
>> >
>> >
>> >
>> >
>> > On 12/4/11 10:22 AM, "kirtana S" <skirtana4.gmail.com> wrote:
>> >
>> > >I require to use non bonded interaction of ion in my monomer . This
>>was
>> > >the
>> > >reason for the difference in energetics. Without this non bonded
>> > >interaction the energetics are good. I used low temperature
>> equilibration
>> > >and with a heating with a time step of 0.0005 ps.
>> > >
>> > >SANDER BOMB in subroutine nonbond_list
>> > >  volume of ucell too big, too many subcells
>> > >  list grid memory needs to be reallocated, restart sander
>> > >
>> > >Can anyone suggest how should I go about to have a non bonded ioninc
>> > >interaction in each monomer group.
>> > >
>> > >Thanks
>> > >Kirtana
>> > >On Sun, Dec 4, 2011 at 3:42 AM, Yong Duan <duan.ucdavis.edu> wrote:
>> > >
>> > >>
>> > >> Interesting. Do you have an explanation on why the energies are so
>> > >> different?
>> > >>
>> > >> Looks like you have a poorly equilibrated system. You may want to
>>try
>> > >> short equilibration at low temperature.
>> > >>
>> > >> What time step did you use? Any other relevant information you can
>> > >>provide?
>> > >>
>> > >> --
>> > >> Yong Duan, Ph.D, Professor
>> > >> UC Davis Genome Center and
>> > >> Department of Biomedical Engineering
>> > >> University of California at Davis
>> > >> Davis, CA 95616
>> > >> 530-754-7632
>> > >>
>> > >>
>> > >>
>> > >>
>> > >>
>> > >>
>> > >> On 12/3/11 4:02 PM, "kirtana S" <skirtana4.gmail.com> wrote:
>> > >>
>> > >> >Dear All,
>> > >> >
>> > >> >I ran minimization of the solvent, minimization of entire system ,
>> > >> >constant
>> > >> >volume equilibration and again minimization before the constant
>> > >>pressure
>> > >> >equilibration.
>> > >> >The job gets killed after this step with an error as below, is
>>this
>> > >> >something to do with my input prmtop and inpcrd files
>> > >> >
>> > >> >vlimit exceeded for step      1; vmax =   486.1523
>> > >> >vlimit exceeded for step      2; vmax = **********
>> > >> >vlimit exceeded for step      3; vmax = **********
>> > >> >vlimit exceeded for step      4; vmax = **********
>> > >> >vlimit exceeded for step      5; vmax = **********
>> > >> >vlimit exceeded for step      6; vmax = **********
>> > >> >vlimit exceeded for step      7; vmax = **********
>> > >> >vlimit exceeded for step      8; vmax = **********
>> > >> >vlimit exceeded for step      9; vmax = **********
>> > >> >vlimit exceeded for step     10; vmax = **********
>> > >> > SANDER BOMB in subroutine nonbond_list
>> > >> >  volume of ucell too big, too many subcells
>> > >> >  list grid memory needs to be reallocated, restart sander
>> > >> >
>> > >> >The last step of  minimization has
>> > >> > FINAL RESULTS
>> > >> >
>> > >> >   NSTEP       ENERGY          RMS            GMAX         NAME
>> > >>NUMBER
>> > >> >   1000      -1.2257E+05     9.7831E-01     1.2756E+02     C7
>> > >>551
>> > >> > BOND    =     6004.1653  ANGLE   =     1318.8747  DIHED      =
>> > >> >205.2874
>> > >> > VDWAALS =    16210.2078  EEL     =  -153314.2089  HBOND      =
>> > >> >0.0000
>> > >> > 1-4 VDW =      268.8998  1-4 EEL =     6739.7131  RESTRAINT  =
>> > >> >0.0000
>> > >> >and equilibration file has
>> > >> >NSTEP =        0   TIME(PS) =       0.000  TEMP(K) =   244.40
>>PRESS
>> =
>> > >> >2013.9
>> > >> > Etot   =    -17407.5244  EKtot   =     18506.4151  EPtot      =
>> > >> >-35913.9396
>> > >> > BOND   =     12704.4067  ANGLE   =      1455.1636  DIHED      =
>> > >> >0.3601
>> > >> > 1-4 NB =       541.0429  1-4 EEL =      6726.3304  VDWAALS    =
>> > >> >5828.9860
>> > >> > EELEC  =    -63170.2293  EHBOND  =         0.0000  RESTRAINT  =
>> > >> >0.0000
>> > >> > EKCMT  =      6364.7776  VIRIAL  =    -20026.3462  VOLUME     =
>> > >> >606942.3409
>> > >> >                                                    Density    =
>> > >> >0.3860
>> > >> > Ewald error estimate:   0.2325E-03
>> > >> >
>> > >>
>> >
>> 
>>>>>----------------------------------------------------------------------
>>>>>--
>> > >>>--
>> > >> >----
>> > >> >vlimit exceeded for step      1; vmax =   486.1523
>> > >> >vlimit exceeded for step      2; vmax = **********
>> > >> >
>> > >> >Thanks for suggestion
>> > >> >Kirtana
>> > >> >On Sat, Dec 3, 2011 at 6:07 PM, case <case.biomaps.rutgers.edu>
>> wrote:
>> > >> >
>> > >> >> On Sat, Dec 03, 2011, kirtana S wrote:
>> > >> >> >
>> > >> >> > While running constant pressure equilibration after
>>minimization
>> I
>> > >>get
>> > >> >> the
>> > >> >> > following error
>> > >> >> >
>> > >> >> > vlimit exceeded for step      1; vmax =   365.9801
>> > >> >> > vlimit exceeded for step      2; vmax = **********
>> > >> >>
>> > >> >> You should generally equilibrate at constant volume first, and
>>then
>> > >>turn
>> > >> >> on constant pressure.  I note that you density is very low, but
>> maybe
>> > >> >>that
>> > >> >> is what you expect here ("solvent box of monomers").  If not,
>>check
>> > >>to
>> > >> >>see
>> > >> >> what is going on.
>> > >> >>
>> > >> >> The fact that such bad things happen on the first two steps may
>> have
>> > >> >> nothing
>> > >> >> to do with constant pressure.  Be sure that the potential energy
>> > >>starts
>> > >> >>at
>> > >> >> the
>> > >> >> same place it had at the end of the minimization.
>> > >> >>
>> > >> >> ....dac
>> > >> >>
>> > >> >>
>> > >> >> _______________________________________________
>> > >> >> AMBER mailing list
>> > >> >> AMBER.ambermd.org
>> > >> >> http://lists.ambermd.org/mailman/listinfo/amber
>> > >> >>
>> > >> >_______________________________________________
>> > >> >AMBER mailing list
>> > >> >AMBER.ambermd.org
>> > >> >http://lists.ambermd.org/mailman/listinfo/amber
>> > >>
>> > >>
>> > >>
>> > >> _______________________________________________
>> > >> AMBER mailing list
>> > >> AMBER.ambermd.org
>> > >> http://lists.ambermd.org/mailman/listinfo/amber
>> > >>
>> > >_______________________________________________
>> > >AMBER mailing list
>> > >AMBER.ambermd.org
>> > >http://lists.ambermd.org/mailman/listinfo/amber
>> >
>> >
>> >
>> > _______________________________________________
>> > AMBER mailing list
>> > AMBER.ambermd.org
>> > http://lists.ambermd.org/mailman/listinfo/amber
>> >
>>
>>
>>
>> --
>> ================================ Current Address =======================
>>  Brian Radak                                             :     BioMaPS
>> Institute for Quantitative Biology
>>  PhD candidate - York Research Group       :     Rutgers, The State
>> University of New Jersey
>>  University of Minnesota - Twin Cities         :     Center for
>>Integrative
>> Proteomics Room 308
>>  Graduate Program in Chemical Physics     :     174 Frelinghuysen Road,
>>  Department of Chemistry                          :     Piscataway, NJ
>> 08854-8066
>>  radak004.umn.edu                                 :
>> radakb.biomaps.rutgers.edu
>> ====================================================================
>> Sorry for the multiple e-mail addresses, just use the institute
>>appropriate
>> address.
>>  _______________________________________________
>> AMBER mailing list
>> AMBER.ambermd.org
>> http://lists.ambermd.org/mailman/listinfo/amber
>>
>_______________________________________________
>AMBER mailing list
>AMBER.ambermd.org
>http://lists.ambermd.org/mailman/listinfo/amber
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Sun Dec 04 2011 - 13:00:03 PST
Custom Search