[Biopython-dev] plink phasing and others
Peter Cock
p.j.a.cock at googlemail.com
Mon Apr 16 16:08:20 UTC 2012
2012/4/16 Tiago Antão <tiagoantao at gmail.com>:
> 2012/4/16 Peter Cock <p.j.a.cock at googlemail.com>:
>> For simple Unix style applications controlled via the command line,
>> use the Bio.Application framework as in Bio.Align.Applications or
>> Bio.Sequencing.Applications, Bio.Phylo.Applications, or
>> Bio.Emboss.Applications (etc?).
>
> I wonder if people never had the need to abstract the computing
> infrastructure? The current code does local (blocking) execution, but
> we see environments with BAS or grids where other models are used. I
> am not suggesting any specific solution, but the current approach
> seems to me not very scalable. No?
I use the current framework with an SGE cluster, str(cline_object)
gives the command line string to submit as the jobs.
It would be nice to have some documented examples using
this in combination with multiprocessing or something... but
I find most of the tools I call are already multi-threaded.
Peter
More information about the Biopython-dev
mailing list