[gmx-users] avoiding bad contact in continutation run

[Justin A. Lemkul]

Adwait Mevada wrote:
>>    1. Re: avoiding bad contact in continutation run (Mark    Abraham)
>> Adwait Mevada wrote:
>>>
>>>> On 01/20/11, Adwait Mevada  <adwait at imsc.res.in> wrote:
>>>>>
>>>>>  Dear gmx-users,
>>>>>  I am simulating a dppc + chol system, the force field is martini
>>>>> with gromacs 3.3.3 the system was initially run for 150 ns, but i
>>>>> had to later extend the simulation, so using the confout.gro and
>>>>> with mdp options:
>>>>> gen_vel = no
>>>>> unconstrained_start = yes
>>>>>  I gave a re-run for another 850 ns, now due to some interaction  the
>>>>> box explodes after 460ns of run.I found that the cause was due  to
>>>>> some unwanted bad contact.
>>>>> It is important that the system continue from the previous state,
>>>>> but the extension is leading to this bad contact.
>>>>>
>>>>
>>>> Actually, you didn't preserve your ensemble if you used the .gro  file
>>>> for the coordinates. See
>>>> http://www.gromacs.org/Documentation/How-tos/Doing_Restarts Actually
>>>>  providing us with copies of the commands you're using is always a
>>>> good idea when asking for help.
>>>
>>> 150ns.gro was the output of the previous simulation run i.e. confout.gro
>>> mdp file for the previous simulation is taken as is except for change of
>>> simulation steps and making unconstrained_start = yes
>>> grompp -np 8 -f 850ns.mdp -p dppc_chol.top -c 150ns.gro -o
>>> 512lip_256chol_850ns.tpr
>>
>> As Mark suspected, you have failed to preserve your ensemble, due to 
>> the fact
>> that you're using a reduced precision (.gro) format for coordinates, 
>>  and you are
>> not supplying the .trr and .edr file from the previous run, so you're 
>> losing
>> velocities, temperature and pressure coupling information, etc.
>>
>> -Justin
> so i should be doing something like
> tpbconv -s 512lip_150ns.tpr -f 512lip_150ns.trr -e 512lip_150ns.edr -o 
> 512lip_850ns.tpr -extend 850
> 
> mpiexe.openmpi -np 8 mdrun_mpi.openmpi -np 8 -s 512lip_850ns.tpr
>  -o 512lip_850ns -e 512lip_850ns -g 512lip_850ns &
> 

Yes, that would be an appropriate set of commands to extend the run.

-Justin

>>> mpiexe.openmpi -np 8 mdrun_mpi.openmpi -np 8 -s 512lip_256chol_850ns.tpr
>>> -o 512lip_256chol_850ns -e 512lip_256chol_850ns -g 
>>> 512lip_256chol_850ns &
>>>
>>>> Mark
>>>>
>>>>>
>>>>> My question is, is there a way to avoid this while retaining the
>>>>> continuation of the previous simulation?
>>>>>  Is using lower precision file a solution to this problem? if yes
>>>>> how do i go about it.
>>>>> -Adwait
>>>>>
>>>>> ----------------------------------------------------------------
>>>>> This message was sent using IMP, the Internet Messaging Program.
>>>>>
>>>>> -- 
> 
> 
> -Adwait
> 
> ----------------------------------------------------------------
> This message was sent using IMP, the Internet Messaging Program.
> 

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

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

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