From: Jérôme Hénin (jerome.henin_at_ibpc.fr)
Date: Fri Jun 26 2020 - 07:19:35 CDT
I don't think NAMD offers the option to append to the output file. Definitely not to the fepout file anyway. You might have to truncate the data from the interrupted window at the end of the fepout file.
Ideally, the job should be designed so that it runs complete FEP windows and terminates before it gets killed by the cluster.
Best,
Jérôme
----- On 25 Juin 20, at 14:12, Jérôme Hénin jerome.henin_at_ibpc.fr wrote:
> Dear Qasim,
>
> To restart the simulation, you should find the last lambda value that was
> computed. If it was, for example, 0.7, then the lines at the end of the config
> file would become:
> runFEP 0.7 1.0 0.1 $nSteps
>
> That is, run from lambda=0.7 to 1, in increments of 0.1.
>
> And then you continue with the backwards simulation:
> runFEP 1.0 0.0 -0.1 $nSteps
>
> which may also end up split into several NAMD runs.
>
> Best,
> Jerome
>
> ----- On 25 Juin 20, at 9:37, Qasim Pars qasimpars_at_gmail.com wrote:
>
>> Dear users,
>>
>> I run a free energy simulation as in this tutorial
>> https://www.ks.uiuc.edu/Training/Tutorials/namd/ABF/ABF-tutorial-files/Methane-Hydration/FEP/decouple-forward.conf
>> but my simulation stops due to the time limit of slurm system. Do you know
>> I can restart the simulation?
>> It would be good if you write the restarting command.
>>
>> Regards,
>> --
> > Qasim Pars
This archive was generated by hypermail 2.1.6 : Thu Dec 31 2020 - 23:17:13 CST