[gmx-users] Peptide - DMPC membrane simulations -> Unstable system

[Justin A. Lemkul]

Kirill Bessonov wrote:
> Hello gain Justin,
> 
> Now I figured out that the problem was in gromacs version 4.0. So 
> peptide in water works no problem.
> 
> I have used trjconv -pbc mol -ur compact and it solved the artifact 
> problem. Thanks a lot!
> 
> Now I have started simulation on 96 CPU's for 1000ns total run time. But 
> after 1 day simulation the system crashed or exploded with the following 
> error: [First time it failed because of some problems with the server, 
> and then I have resumed the run using tpbconv -s DMPCRun1000ns.tpr -f 
> 1000nsDMPCA141Run_part1.trr -extend 100000 -o DMPCRun100nspart2.tpr]
> 

Which simulation?  Peptide in water, or peptide-DMPC?  This will be important 
regarding the .mdp file you use.

Also, this use of tpbconv is obsolete.  Use checkpoint files to obtain an exact 
restart.

> vol 0.78  imb F  2% pme/F 2.15 step 10578400, will finish Sat Aug 29 
> 07:23:01 2009
> vol 0.77  imb F  3% pme/F 2.15 step 10578500, will finish Sat Aug 29 
> 07:28:32 2009
> 
> Step 10578599, time 21157.2 (ps)  LINCS WARNING
> relative constraint deviation after LINCS:
> rms 0.299801, max 6.073617 (between atoms 9449 and 9450)
> bonds that rotated more than 30 degrees:
>  atom 1 atom 2  angle  previous, current, constraint length
>    9451   9468   93.6    0.1530   0.1293      0.1530
> 
> Fatal error:
> 15 particles communicated to PME node 6 are more than a cell length out 
> of the domain decomposition cell of their charge group
> 
> I am attaching my mdp files used to to EM and MDrun here:
> 
> EM: http://kbessonov.googlepages.com/em_lipid.mdp MD: 
> http://kbessonov.googlepages.com/mdonly_1000ns.mdp
> 
> I was wondering why my system fails. and what can I do to avoid it? I 
> was thinking maybe to run several times EM on the system, maybe it will 
> help? What bugs me is that it fails really late in simulation. I have 
> not seen trajectory file yet.
> 

I'm guessing this is for your lipid system (based on the em.mdp file name)? 
What kind of equilibration are you using?  Any position-restrained dynamics?  If 
you're not, then you probably should be :)

Any other comments are subject to knowing what system you're really dealing 
with.  If this is a peptide-membrane system, you'll need to make alterations to 
your handling of COM motion removal, temperature coupling, and pressure coupling.

-Justin

> 
> 
> ------------------------------------------------------------------------
> 
> _______________________________________________
> gmx-users mailing list    gmx-users at gromacs.org
> http://lists.gromacs.org/mailman/listinfo/gmx-users
> Please search the archive at http://www.gromacs.org/search before posting!
> Please don't post (un)subscribe requests to the list. Use the 
> www interface or send it to gmx-users-request at gromacs.org.
> Can't post? Read http://www.gromacs.org/mailing_lists/users.php

-- 
========================================

Justin A. Lemkul
Ph.D. Candidate
ICTAS Doctoral Scholar
Department of Biochemistry
Virginia Tech
Blacksburg, VA
jalemkul[at]vt.edu | (540) 231-9080
http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin

========================================