Hi Robert,
Upon looking into this further I think the issue is with the manual being
somewhat vague about what nstep0 actually does. It would appear that this
value only affects the effective counter of steps for the NMR restraints. So
for example if you had the following input:
&wt type='NSTEP0', istep1=10 /
&wt type='TEMP0', istep1=0, istep2=10,
value1=0.0, value2=30.0 /
&wt type='TEMP0', istep1=11, istep2=20,
value1=30.0, value2=0.0 /
&wt type='END' /
This would mean that on the first step of your calculation (whether or not
it would be a restart) the value of temp0 would be set to 30.0 and not 0.0.
This is really just for convenience in restarts and would only be needed if
you are using NMR restraints to ramp the temperature for example. You could
of course write the following as:
&wt type='TEMP0', istep1=0, istep2=10,
value1=30.0, value2=0.0 /
&wt type='END' /
Which would have the same effect as setting nstep0 = 10 and so skipping the
first 10 steps of ramping. Thus nstep0 does not affect the actual printing
of nstep in the output file - this always starts from 0 (1 for a restart)
and runs to nstlim. The time however is read from a restart so really this
is the column you would want as you x axis in anything you were plotting.
I hope this helps. I will tweak the manual entry for AMBER 11 so this is not
as confusing.
All the best
Ross
> -----Original Message-----
> From: amber-bounces.ambermd.org [mailto:amber-bounces.ambermd.org] On
> Behalf Of Ross Walker
> Sent: Wednesday, February 04, 2009 7:04 PM
> To: 'AMBER Mailing List'
> Subject: RE: [AMBER] Restarting an MD Simulation
>
> Hi Robert,
>
> To be honest I didn't even know this option existed and I'm not surprised
> that it doesn't work. When I get back online I'll try this myself and see
> if
> I can reproduce it. One thing to note though is that there is nothing
> actually wrong with your restarted MD run except that the value of nstep
> got
> reset. Everything else including the Time should continue as if it were
> restarted. Hence if you were to plot things against the time in the output
> file it should be continuous. Also reading in the two trajectories into
> ptraj for example, one after the other, should give you the same results
> as
> if it were a single run.
>
> It would probably be possible to devise some kind of awk script that would
> go through and add 179000 to every nstep value in your output file which
> would give you what you were expecting I guess.
>
> All the best
> Ross
>
> > -----Original Message-----
> > From: amber-bounces.ambermd.org [mailto:amber-bounces.ambermd.org] On
> > Behalf Of Hopkins, Robert
> > Sent: Wednesday, February 04, 2009 4:34 PM
> > To: AMBER.ambermd.org
> > Subject: [AMBER] Restarting an MD Simulation
> >
> > Amber Users,
> >
> >
> >
> > I'm using AMBER9 to do some MD calculations and one of my runs was
> > stopped after NSTEP = 179000 because of a power glitch. The restart
> > file at this point was named IIL.rst. After reading about Weight
> > Change Information for NSTEP0 on p. 109 of the User's Manual, it
> > appeared that restarting the run with NSTEP=180000 and continuing to
> > NSTEP=250000 should be straightforward. So, I used the following mdin
> > file:
> >
> >
> >
> > Continue dodecamer equilibration at 300K & P=const.; no DNA restraints;
> >
> > SHAKE on; MD on; from NSTEP 180000 to 250000.
> >
> > &cntrl
> >
> > imin = 0, irest = 1, ntx = 5,
> >
> > nstlim = 71000, dt = 0.002,
> >
> > ntc = 2, ntf = 2,
> >
> > cut = 8.0, ntb = 2, ntp = 1, taup = 2.0,
> >
> > ntpr = 1000, ntwx = 1000,
> >
> > ntt = 3, gamma_ln = 2.0,
> >
> > temp0 = 300.0, nmropt =1
> >
> > /
> >
> > &wt type='NSTEP0', istep1=179000 /
> >
> > &wt type='END' /
> >
> >
> >
> > along with the command:
> >
> >
> >
> > sander -O -i mdin -o IIL2.out -p IIL_sol.prmtop -c IIL.rst -r
> > IIL2.rst -x IIL2.mdcrd &
> >
> >
> >
> > The MD calculation evidently proceeded as expected except that the
> > values of NSTEP began at 1000 and continued to 71000. I'd appreciate
> > any suggestions of what I need to do differently to correct this result.
> > Thanks.
> >
> >
> >
> > Bob Hopkins
> >
> > _______________________________________________
> > 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 Fri Feb 06 2009 - 01:15:55 PST
