Hello. Thank you for the reply. In fact, a time step of 1fs seems to
work around the problem. At the same time as i was writing the last
email, I had submitted a job with dt = 0.001 and it finished without
any problems. I haven´t tested yet with pmemd.cuda, but I´ll do that
later on.
The two atoms involved in this case are O and H of the hydroxyl
residue linked to the zinc atom of the active site. Where do I find
the Lennard-Jones rmin values for them? The hole active site was
parameterized following the steps on the MTK++ manual.
I´ll follow your instructions as to getting a restart file from a
previous step and test it out. I´ll inform you as soon as I have the
results.
Thank you
Fabrício Bracht
2012/6/28 David A Case <case.biomaps.rutgers.edu>:
> On Thu, Jun 28, 2012, Fabrício Bracht wrote:
>
>> Hi. I did what you asked. I ran the simulation again, in order to make
>> sure that the error happened again. This time, it dies a lot earlier
>> with the same warnings as last time (vlimit exceed....etc). I
>> continued the simulation using as input the last restrt file written
>> using the serial version of pmemd. The simulation crashed after a few
>> hours , the last output lines are:
>>
>>
>>
>> vlimit exceeded for step 112126; vmax = 91.1788
>> vlimit exceeded for step 112176; vmax = 45.6806
>>
>> Coordinate resetting cannot be accomplished,
>> deviation is too large
>> iter_cnt, my_bond_idx, i and j are : 2 2690 5315 5316
>>
>
> Thanks for the info. It looks to me like the problem has nothing to do (per
> se) with GPU's, since you can reproduce it on a serial CPU using pmemd.
> Unfortuantely, problems like this (generally with the force field) can be hard
> to track down. Your mdin file looks fine to me, although it is safer to use
> dt=0.001 rather than 0.002, especially for constant pressure simulations, and
> if you system is not *quite* equilibrated.
>
> The information above indicates that the problem occurs around atoms 5315
> and 5316. You might see what those are, as this might (or might not) give
> a clue to the origin of the problem. Look hard (and/or let us know about)
> any non-standard residues or molecules that are in your simulation. Do
> you have any Lennard-Jones rmin values that are either zero or small (i.e.
> less than typical values for that type of atom)? You may have to get a
> restart file someplace a little before step 112126; (don't use ig=-1 to do
> this: copy the actual seed that is printed in your output file, so you will
> get an identical trajectory.) Then you can set ntwx=1, and examine the
> coordinates at every step as the system crashes....one hopes that this will
> identify the source of the problem.
>
> ...good luck...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
Received on Thu Jun 28 2012 - 12:30:02 PDT
