Hi!
Yes, you are right, I have never encountered a problem with aMD and, the
way that the code works, it should be pretty stable.
It seems the problem has been resolved. I got an email fro John saying he
updated his local version and seems to be working now. I will check his
files nonetheless just to be sure.
Thanks to both for comments and for sending possible problems when you
find them :)
Best,
Romelia
> I had just hit a vlimit (velocity limit) issue with a non-AMD simulation I
> was working on.
>
> For my work, an otherwise ligand held in place via an NMR restraint to
> molecule 1 was imaged to an adjoing cell, so on restart, there was 70,000
> kCal of energy where there had only been 10 kCal in the closing restart
> file of the prior run.
>
> This resulted in tremendous velocity being transferred to the restrained
> ligand - and the vlimit was absolutely exceeded.
>
> I have not had this problem with an AMD run (yet). Mathematically, amd
> adds to the force calculations in a quite simple way, so I would be
> surprised to find AMD specifically to be at the root of the problem.
> Review the energetics of the coordinates under simulation first. Are
> there any high energy terms?
>
> If I were debugging this from computer-scientist perspective, I would dial
> down the boosting potentials so that there is in essence "no boost" - and
> review my settings of those parameters extra-carefully. Are temperature
> controls being performed in a manner consistent with the recommendations
> of the AMD community?
>
> ________________________________________
> From: Beale, John [John.Beale.stlcop.edu]
> Sent: Thursday, February 28, 2013 7:27 AM
> To: amber.ambermd.org
> Subject: [AMBER] aMD
>
> Hello AMBER users,
>
> I am trying to run an accelerated MD simulation on a 7-residue peptide in
> explicit water (1234 waters). I have followed the protocol in the Amber-12
> manual exactly, choosing a small peptide molecule to do my first
> simulations. My input file is below (this is the sample input given on
> pages 160-161 in the Amber-12 manual, with my own parameters for EthreshD,
> alphaD, EthreshP, and alphaP).
>
> aMD dt=2.0fs with SHAKE, NPT AMD boost pot and dih
> &cntrl
> imin = 0, irest = 1, ntx = 5,
> dt=0.002, ntc=2, ntf=2, tol=0.000001, iwrap=1,
> ntb=2, cut=12.0, ntp=1, igb=0, ioutfm = 1,
> ntt = 3, temp0=310.0, gamma_ln=1.0, ig = -1,
> ntpr = 1000, ntwx = 1000, ntwr = 1000, nstlim = 2000000,
> iamd = 3, EthreshD=85,
> alphaD = 5, EthreshP = -11180,
> alphaP = 765,
> /
> &ewald
> dsum_tol=0.000001,
> /
>
> My first step was to run a normal dynamics simulation on the equilibrated
> system, and from this I calculated the aMD parameters. From this, my
> average dihedral was 60.4693, average EPtot was -11945.7808, total ATOMS
> was 3827, and protein residue number is 7.
>
> Using pmemd.MPI I have made numerous attempts to try an aMD simulation on
> this system. Each time the run stops and I get a string of error messages
> that tell me that vlimit was exceeded. Can someone tell me what this
> means? What should I be doing to correct this so the simulation will run.
> I would appreciate any suggestions that anyone might give me.
>
> Thanks!
>
> John
>
> John M. Beale, Jr., Ph.D.
> Professor of Medicinal Chemistry and Pharmacognosy
> Saint Louis College of Pharmacy
> 4588 Parkview Place
> Saint Louis, Missouri 63110-1088
> Office: 314-446-8461
> Cell: 314-315-0409
> FAX: 314-446-8460
> John.Beale.stlcop.edu<mailto:John.Beale.stlcop.edu>
>
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
--
****************************************
Romelia Salomon
Walker Group
398 San Diego Supercomputing Center
UC San Diego
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Thu Feb 28 2013 - 08:30:05 PST
