[Biopython] Shell permission denied

Justin Gibbons jgibbons1 at mail.usf.edu
Tue Jul 30 16:01:30 UTC 2013


Since its working from the command line the first thing I would try is
using the subprocess <http://docs.python.org/2/library/subprocess.html>module
instead of os.system().

Hope that helps,

Justin Gibbons


On Tue, Jul 30, 2013 at 8:15 AM, Peter Cock <p.j.a.cock at googlemail.com>wrote:

> On Tue, Jul 30, 2013 at 1:04 PM, Abel Valenzuela García
> <avalgar at hotmail.com> wrote:
> > Dear all,
> >
> >
> > I'm using Python 2.7.3 under Ubuntu 12.04 (precise pangolin). My best
> guess is that this has to do with the linux system, or its relationship
> with Python; it's very unlikely that the code is faulty.
> >
> > At some point of my script execution, there is a system call to run a
> program from the linux shell that looks like this:
> >
> > os.system("%s %s > %s" % (DSSP, in_file, out_file.name))
> >  This should basically run the command line
> >
> > DSSP in_file > out_file
> >
> > Here is the source code
> >
> >
> >
> > The ERROR message I get (excerpt from my session):
> >
> > In [8]: p = PDBParser()
> > In [9]: structure = p.get_structure("4E4Z", "4E4Z.pdb")
> > In [10]: model = structure[0]
> > In [11]: dssp = DSSP(model, "4E4Z.pdb")
> > sh: 1: dssp: Permission denied
> >
> > I followed the class documentation for that example, have
> >  a sane pdb file, a dssp package that works nicely and produces correct
> > output from the command line, all permissions to execute, and I'm the
> only user.
> >
> >
> > Any ideas why this might not be working?
> >
> >
> > Thank you very much for you patience and help!
> >
> >
> > Abel Valenzuela
>
> Hi Abel,
>
> In this kind of situation the first thing I do is work out what
> the command line that Python is trying to run is (maybe
> you can add some print statements to the DSSP code?),
> and then try to run that exact same command by hand
> at the terminal.
>
> Another thing to watch out for is spaces in filenames -
> the can be dealt with using quotes or escaping, but
> sometimes this defensive coding hasn't been done.
>
> Perhaps we need some more unit tests for this part
> of Biopython?
>
> Peter
>
> _______________________________________________
> Biopython mailing list  -  Biopython at lists.open-bio.org
> http://lists.open-bio.org/mailman/listinfo/biopython
>




More information about the Biopython mailing list