Re: [AMBER] Sander doing the minimisation but stopping at the first heating step, not generating rst or nc files

From: Adrian Roitberg <roitberg.ufl.edu>
Date: Thu, 20 Jul 2017 15:04:35 -0400

Hi

If you have a trajectory file, you can look at it and see if you find
something strange.

If you sent this job to a queue, then you might get an error message
from data, which you can check.

As for the temperature: unless your system was fully minimized, at step
1 you would have forces in the system, which would move the atoms,
impose a velocity, and translate that into kinetic energy (aka temperature)

Now, I am 100% sure your system was not minimized properly, because your
initial bond energy is very large, and at step 40 it becomes much lower.
Your restraint energy is VERy hugh, which tells me that by step 40, you
are not really close to whatever reference structure you choose to
restrain against.

Adrian


On 7/20/17 2:58 PM, Alonso Martinez, Diego wrote:
> Dear all,
>
> Many thanks for your inputs.
>
> I’ve followed your recommendations. The simulation now runs up to 40 steps where it stops. The output for step 0 is:
>
> NSTEP = 0 TIME(PS) = 0.000 TEMP(K) = 0.00 PRESS = 0.0
> Etot = -377171.9116 EKtot = 0.0000 EPtot = -377171.9116
> BOND = 22001.3894 ANGLE = 5591.1333 DIHED = 11036.2664
> 1-4 NB = 4130.6427 1-4 EEL = 25899.8412 VDWAALS = 22638.7966
> EELEC = -468469.9812 EHBOND = 0.0000 RESTRAINT = 0.0000
> Ewald error estimate: 0.9785E-03
> ------------------------------------------------------------------------------
>
> NMR restraints: Bond = 0.000 Angle = 0.000 Torsion = 0.000
>
>
> And the output for the last step (40) is:
>
> NSTEP = 40 TIME(PS) = 0.080 TEMP(K) = 158.88 PRESS = 0.0
> Etot = -382330.0469 EKtot = 45123.8992 EPtot = -427453.9461
> BOND = 8652.6936 ANGLE = 4998.5073 DIHED = 11097.1310
> 1-4 NB = 3798.8621 1-4 EEL = 25049.1304 VDWAALS = 47662.3745
> EELEC = -542003.8679 EHBOND = 0.0000 RESTRAINT = 13291.2230
> EAMBER (non-restraint) = -440745.1691
> Ewald error estimate: 0.5425E-03
> ------------------------------------------------------------------------------
>
> NMR restraints: Bond = 0.000 Angle = 0.000 Torsion = 0.000
>
>
> What surprised me the most about this outcome is the fact that I set the temperature to increase from 0 to 100 K but it does increase up to 158.88K. Do you have any input about why this could be happening? Many thanks.
>
> Best wishes,
> Diego
>
>
> On 20/07/2017, 14:40, "Carlos Simmerling" <carlos.simmerling.gmail.com> wrote:
>
> also run with single thread, not MPI.
>
> On Thu, Jul 20, 2017 at 9:22 AM, Adrian Roitberg <roitberg.ufl.edu> wrote:
>
> > Try setting ntpr=1 and ntwx=1 and see if you ever go beyond the first
> > step and what happens.
> >
> > Right now you have them set to 5000, so lots of things could have
> > happened between those steps.
> >
> > adrian
> >
> >
> > On 7/20/17 9:05 AM, Alonso Martinez, Diego wrote:
> > > Dear Amber users,
> > >
> > > I am trying to obtain a MD trajectory of a DNA-protein-metal-ligand
> > complex.
> > >
> > > To prepare the inpcrd and prmtop files, I followed the following
> > approach: first, I’ve prepared PDB files of each of the elements of the
> > simulation: the protein and the DNA with the H++ webserver, the ligand and
> > metal with antechamber. Then, I joined all the pdb files into one with the
> > cat command, to then generate the inpcrd and prmtop files of the complex
> > with tleap.
> > >
> > > The tleap input file was as follows:
> > >
> > > source leaprc.protein.ff14SB
> > > source leaprc.DNA.OL15
> > > source leaprc.water.tip3p
> > > source leaprc.gaff2
> > > CPF = loadmol2 CPF.mol2
> > > loadamberparams CPF.frcmod
> > > loadamberparams frcmod.ions234lm_126_tip3p
> > > mol = loadpdb Gyrase_complex_fixed.pdb
> > > savepdb mol Gyrase_complex_dry.pdb
> > > saveamberparm mol Gyrase_complex_dry.prmtop Gyrase_complex_dry.inpcrd
> > > solvateoct mol TIP3PBOX 15.0
> > > addions mol Na+ 0
> > > addions mol Cl- 0
> > > savepdb mol Gyrase_complex_solv.pdb
> > > saveamberparm mol Gyrase_complex_solv.prmtop Gyrase_complex_solv.inpcrd
> > > quit
> > >
> > > The method used to parametrize the metal ion is the nonbonded model as
> > we want to observe if the metal ion exits the binding site upon mutations
> > in the protein. The parametrization was checked with parmed.
> > >
> > > Inpcrd and prmtop files of the solvated complex were then generated.
> > Then, we proceeded with an energy minimization using sander.MPI, followed
> > by heating with sander.MPI as well. The minimization run fine, at least
> > when compared to previously successful minimizations. However, when it
> > passed to the heating, the mdout file only prints the first step and does
> > not continue the simulation after that, not even generating a trajectory
> > file. The mdout file was:
> > >
> > > NSTEP = 0 TIME(PS) = 0.000 TEMP(K) = 0.00 PRESS =
> > 0.0
> > > Etot = -393779.1731 EKtot = 0.0000 EPtot =
> > -393779.1731
> > > BOND = 23619.4348 ANGLE = 5814.7119 DIHED =
> > 11047.5656
> > > 1-4 NB = 4193.6173 1-4 EEL = 25963.2849 VDWAALS =
> > 29449.7359
> > > EELEC = -493867.5237 EHBOND = 0.0000 RESTRAINT =
> > 0.0000
> > > Ewald error estimate: 0.7721E-03
> > > NMR restraints: Bond = 0.000 Angle = 0.000 Torsion =
> > 0.000
> > > ============================================================
> > ===================
> > >
> > > After which no more steps are printed. The program just exits and
> > doesn’t generate trajectory or restart files to continue to the next
> > heating step. The heating input file was:
> > >
> > > &cntrl
> > > imin=0,
> > > irest=0,
> > > ntx=1,
> > > nstlim=250000, dt=0.002,
> > > ntc=2, ntf=2,
> > > tol=0.0000001,
> > > cut=10.0, ntb=1,
> > > ntpr=5000, ntwx=5000,
> > > ntt=3, gamma_ln=1.0,
> > > ig=-1,
> > > ntr=1,
> > > ntp=0,
> > > ioutfm=1,
> > > nmropt=1,
> > > ntxo=2,
> > > /
> > > &wt TYPE='TEMP0', istep1=0, istep2=250000,
> > > value1=0.0, value2=100.0 /
> > > &wt TYPE='END' /
> > > Hold protein and ligand fixed
> > > 10.0
> > > RES 1 857
> > > END
> > > END
> > >
> > > I’ve tried fixing this issue by removing the energy constraints and
> > changing sander for pmemd but none of these options work. Could you please
> > let me know if you had any issue like this and if so propose possible
> > solutions? Many thanks for your help.
> > >
> > > Best wishes,
> > > Diego
> > >
> > > _______________________________________________
> > > AMBER mailing list
> > > AMBER.ambermd.org
> > > http://lists.ambermd.org/mailman/listinfo/amber
> >
> > --
> > Dr. Adrian E. Roitberg
> > University of Florida Research Foundation Professor
> > Department of Chemistry
> > University of Florida
> > roitberg.ufl.edu
> > 352-392-6972
> >
> >
> > _______________________________________________
> > 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

-- 
Dr. Adrian E. Roitberg
University of Florida Research Foundation Professor
Department of Chemistry
University of Florida
roitberg.ufl.edu
352-392-6972
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Thu Jul 20 2017 - 12:30:02 PDT
Custom Search