Thanks for the quick response Dr. Simmerling,
The starting structures have been docked ligands in a crystal structure, that I had added back in missing amino acid residues manually (this is one of the likely places for high initial energies I know). I saw your posts, and then tried gradual equilibration, but the same result so far.
What's most puzzling, is that everything runs normal (it's not that high energy starts directly evolve into the crash, I can see in the .mdout that the inital frame quickly gets energy minimized from a high starting value), normal in the mdout, normal in VMD visualization, until it blows up suddenly. Only by setting the output frequency to as high as ntpr=1 was I able to see it. I've copied and pasted an example 4 frames here:
NSTEP = 15978 TIME(PS) = 3781.956 TEMP(K) = 312.53 PRESS = 0.0
Etot = -167400.4611 EKtot = 33940.8242 EPtot = -201341.2853
BOND = 1353.4793 ANGLE = 3694.6867 DIHED = 2338.8599
UB = 0.0000 IMP = 0.0000 CMAP = 299.2965
1-4 NB = 1438.9333 1-4 EEL = 15666.6174 VDWAALS = 23181.8003
EELEC = -249314.9587 EHBOND = 0.0000 RESTRAINT = 0.0000
EKCMT = 0.0000 VIRIAL = 0.0000 VOLUME = 523407.2851 Density = 1.0341
------------------------------------------------------------------------------
NSTEP = 15979 TIME(PS) = 3781.958 TEMP(K) = 321.98 PRESS = 0.0
Etot = -166339.6957 EKtot = 34967.1914 EPtot = -201306.8871
BOND = 1332.6273 ANGLE = 3654.6213 DIHED = 2344.1866
UB = 0.0000 IMP = 0.0000 CMAP = 300.2194
1-4 NB = 1602.3619 1-4 EEL = 15665.8723 VDWAALS = 23153.0389
EELEC = -249359.8148 EHBOND = 0.0000 RESTRAINT = 0.0000
EKCMT = 0.0000 VIRIAL = 0.0000 VOLUME = 523407.2851 Density = 1.0341
------------------------------------------------------------------------------
NSTEP = 15980 TIME(PS) = 3781.960 TEMP(K) = 12807.45 PRESS = 0.0
Etot = 1200379.6146 EKtot = 1390876.8750 EPtot = -190497.2604
BOND = 1331.5817 ANGLE = 3724.0317 DIHED = 2343.1848
UB = 0.0000 IMP = 0.0000 CMAP = 300.3352
1-4 NB = 4362.7163 1-4 EEL = 15659.0224 VDWAALS = 31195.5756
EELEC = -249413.7082 EHBOND = 0.0000 RESTRAINT = 0.0000
EKCMT = 0.0000 VIRIAL = 0.0000 VOLUME = 523407.2851 Density = 1.0341
------------------------------------------------------------------------------
NSTEP = 15981 TIME(PS) = 3781.962 TEMP(K) = NaN PRESS = 0.0
Etot = NaN EKtot = NaN EPtot = **************
BOND = 78051.6147 ANGLE = 5028.6654 DIHED = 2347.0270
UB = 0.0000 IMP = 0.0000 CMAP = 298.6265
1-4 NB = 1615.2760 1-4 EEL = 15623.6489 VDWAALS = **************
EELEC = -249473.1008 EHBOND = 0.0000 RESTRAINT = 0.0000
EKCMT = 0.0000 VIRIAL = 0.0000 VOLUME = 523407.2851 Density = 1.0341
I've included my input and output files as a small attachment (the 1MB limit) in the last email of this thread. If that's at anyone's convenience one can use it to have a try of their own.
I had been running with the vlimit=10 parameter on that solves this issue mostly, but I was pursusing a more permanent fix after seeing the recent posts.
------------------------------------------------------------------
Sender:Carlos Simmerling <
carlos.simmerling.gmail.com>
Sent At:2020 Nov. 20 (Fri.) 16:30
Recipient:Liao <
liaojunzhuo.aliyun.com>; AMBER Mailing List <
amber.ambermd.org>
Subject:Re: [AMBER] vlimit=10 compromise for Amber 20 error: "an illegal memory access was encountered launching kernel kClearForces"?
have you seen the post I made earlier about ff19SB errors? I am not sure of the exact issue, but each time I have checked someone's inputs it has turned out that the structure has high energy contacts and worked fine with careful equilibration. Can you tell us how you obtained the structure, and if any parts needed to be modeled? Also your equilibration protocol may be important so it can help to summarize that.
As Dave suggested, a small example is helpful, particularly if it doesn't include unusual force field parameters or heavy atoms that needed to be modeled.
On Fri, Nov 20, 2020 at 4:18 PM Liao <liaojunzhuo.aliyun.com> wrote:
Continuing on this old thread, using vlimit=10 resolved the issue mostly, but there was still a 2% chance the simulation will crash within 5 ns.
Now, working on a new protein-ligand system, I started out with ff14SB, runs normally as expected (Implicit water, HMR prmtop files used). When I decided to try ff19SB also, the simulation blows up again quite quickly. This time it doesn't give an error and stop, but the energy values have suddenly blown up. As mentioned last time by Dr Case, here I've provided my input files, including mol2 and PDB, preparation scripts and a short output. Maybe someone can rerun on their system, to see what their result is, and what modifications can be made? From another recent thread (
http://archive.ambermd.org/202011/0181.html), Dr Simmerling mentioned that a slow 9-step equilibration protocol solved such issue. Don't know if that may be the ultimate solution to all?
------------------------------------------------------------------
Sender:David A Case <david.case.rutgers.edu>
Sent At:2020 Oct. 15 (Thu.) 07:17
Recipient:Liao <
liaojunzhuo.aliyun.com>; AMBER Mailing List <
amber.ambermd.org>
Subject:Re: [AMBER] vlimit=10 compromise for Amber 20 error: "an illegal memory access was encountered launching kernel kClearForces"?
On Wed, Oct 14, 2020, Liao wrote:
>
>The 1st way is when I switch back to ff14SB (with nothing else changed
>but the random seed), by regenerating the prmtop file, it runs and
>completes normally. But since I’ve got Amber 20, I still want to try
>use the ff19sb.
Can you create a (small?) example that illustrates the problem with
ff19sb? We would need a prmtop, inpcrd and mdin files, plus the actual
command line you used to run the program.
...thanks...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
Received on Sun Nov 22 2020 - 16:30:02 PST
