Hi, Victor
In multiple walker ABMD, all the biasing potentials are identical, so
you can just use the first one umbrella.nc or wt_umbrella.nc to analyze.
Best
Feng
On Wed, Nov 18, 2020 at 4:41 AM Victor Jun Yu Lim <limj.uni-marburg.de>
wrote:
>
> Dear AMBER Community,
>
> I am trying to run ABMD using AMBER16 with WT and independent
> multiple-walker.
> The reason I want to use independent multiple-walker instead of a
> selection algorithm is because our infrastructure is set up such that
> we have different servers with GPU that cannot directly interact with
> each other. (SCP/SSH is required to copy data from one to another).
>
> So I was wondering how can read and write to the same bias file across
> the replicates? Is it enough to take the umbrella.nc file from one
> replica and use it for another replica? Can I run it in parallel with
> 2 replicas reading and writing the same umbrella.nc file?
> If that is possible, how do I analyses the multiple wt_umbrella.nc
> generated for each replicates?
>
> Thanks a lot!
> Victor Lim
>
>
>
> --
> Victor Jun Yu Lim
> PhD Student at the AG Kolb
> Institute of Pharmaceutical Chemistry
> FB 16, Department of Pharmacy
> Philipps-University of Marburg
> Marbacher Weg 6, 35037 Marburg
>
> Website http://kolblab.org/victor.html
>
>
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
--
Feng Pan
PostDoc
Florida State University
Department of Statistics
Email: fpan3.ncsu.edu; fpan.fsu.edu
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Fri Dec 04 2020 - 18:00:02 PST
