Re: Interrupted Molecular Dynamics Simulation

From: Felipe Merino (felipe.merino_at_mpi-dortmund.mpg.de)
Date: Tue Jul 28 2015 - 10:34:17 CDT

Hi,

On 28/07/15 16:51, Purushottam Dixit wrote:
> The simplest would be to output a pdb file from the dcd file using
> catdcd and then using that pdb file as a new starting point.
>
> Purushottam
>
This won't work since you will randomize the velocities and it **won't**
be a proper restart.

> On Tue, Jul 28, 2015 at 10:45 AM, Daniela Rivas
> <dani.rivas.r_at_gmail.com <mailto:dani.rivas.r_at_gmail.com>> wrote:
>
> Hi Chitrak,
>
> That's the problem, I didn't put a restart frequency on my .conf
> file, I forgot. I just set an xst and dcd frequency.
>
> So, there's no way I can use the last trajectory saved on my
> output file?
>
> Thank you,
> Daniela.
>
> 2015-07-28 11:34 GMT-03:00 Chitrak Gupta <chgupta_at_mix.wvu.edu
> <mailto:chgupta_at_mix.wvu.edu>>:
>
> Hi Daniela,
>
> I am slightly surprised that NAMD did not create the restart
> files for you. What did you set your restart frequency as?
>
>
> Chitrak.
>
> On Tue, Jul 28, 2015 at 12:03 AM, Felipe Merino
> <felipe.merino_at_mpi-dortmund.mpg.de
> <mailto:felipe.merino_at_mpi-dortmund.mpg.de>> wrote:
>
> Hi Daniela,
>
> If you don't have the velocities for the configuration
> that you want to restart from (either restart.vel or a
> velocity trajectory) then you have to go from scratch.
>
> Felipe
>
>
> On 28/07/15 02:21, Daniela Rivas wrote:
>> Hi Chitrak,
>>
>> Thank you, but the problem is I can't find those restart
>> files. It seems like I forgot to add that line on my conf
>> file. So, I want to know if it's possible to retake my
>> simulation just using the output.dcd file, that'd be the
>> only way, otherwise I'll have to run my dynamics all over
>> again :(
>>
>> Thanks,
>> Daniela.
>>
>> 2015-07-27 14:29 GMT-03:00 Chitrak Gupta
>> <chgupta_at_mix.wvu.edu <mailto:chgupta_at_mix.wvu.edu>>:
>>
>> Hi Daniela,
>>
>> Yes, you can. In fact it is fairly simple. NAMD
>> should have created the restart files, specifically,
>> the *.res.coor, *.res.vel and *.res.xsc files which
>> contain, respectively, coordinates, velocities, and
>> periodic box information. When you restart your
>> simulation, make sure you read these files (the
>> manual should explain how to do this). Remember to
>> not specify a temperature as this is already being
>> read from the *.vel file (I am assuming a constant
>> temperature run. If that's not what you were doing,
>> refer to the manual on how to restart that). Also, if
>> you want the steps to be numbered from where it last
>> crashed, use the "firsttimestep" option.
>>
>>
>> Hope this helps,
>> Chitrak.
>>
>> On Mon, Jul 27, 2015 at 11:32 AM, Daniela Rivas
>> <dani.rivas.r_at_gmail.com
>> <mailto:dani.rivas.r_at_gmail.com>> wrote:
>>
>> Hi,
>>
>> It's my first time using Namd and I don't know
>> how to continue my interrupted molecular dynamics
>> simulation (electricity went off during the
>> weekend). I have my output.dcd and output.xst
>> files and also have the log file, so I know where
>> to start again, I just don't know how to do it.
>>
>> Is there a way I can retake the simulation from
>> the last point saved in the output?
>>
>> I hope someone can help me :)
>>
>> Thanks,
>> Daniela.
>>
>>
>>
>
> --
> Felipe Merino
> Max Planck Institute for Molecular Physiology
> Department of Structural Biochemistry
> Otto-Hahn-Str. 11
> 44227 Dortmund
> Phone:+49 231 133 2306 <tel:%2B49%20231%20133%202306>
>
>
>
>

-- 
Felipe Merino
Max Planck Institute for Molecular Physiology
Department of Structural Biochemistry
Otto-Hahn-Str. 11
44227 Dortmund
Phone: +49 231 133 2306

This archive was generated by hypermail 2.1.6 : Tue Dec 27 2016 - 23:21:15 CST