[Biopython-dev] Medium/long term plans

Peter Cock p.j.a.cock at googlemail.com
Wed Feb 24 07:43:48 UTC 2010


Hi all,

As mentioned in other recent threads, Brad and I were in Tokyo
earlier this month for the DBCLS BioHackathon 2010 (see
http://hackathon3.dbcls.jp/ for details). While there, we met up
with Michiel for an informal dinner meeting, and discussed some
possible plans for Biopython.

=== Short term action points ===

Seq object equality, see:
http://lists.open-bio.org/pipermail/biopython-dev/2010-February/007351.html

Filenames or handles in SeqIO, AlignIO, etc, see:
http://lists.open-bio.org/pipermail/biopython-dev/2010-February/007352.html

=== Medium term action points ===

Python 3 support. With NumPy starting to make serious plans for
supporting Python 3 this year, we should be able to look at doing this
too. Initially we will continue to focus on Python 2.x, but make more
effort to ensure that we can run without issues in the "Python 3
warning mode" available in Python 2.6 (or 2.7 once that is out).
Then start to put Biopython through 2to3, and see how we get on.

Name space reorganisation for sequences. It would be nice to
have the Seq objects, SeqFeature, SeqRecord and probably
SeqUtils and SeqIO all under one module name. We may be
able to handle this in the short term with two import routes with
the old module names discouraged and eventually deprecated.
See also the "Code review request for phyloxml branch" thread
which covered some of this:
http://lists.open-bio.org/pipermail/biopython-dev/2010-January/007215.html

=== Long term action points ===

There are things in Biopython that with hindsight we feel have not
worked out so well (module naming, alphabets objects) where
change may require a break, i.e. a Biopython version two. Should
we start a wiki to record points of debate, and get people to list their
niggles/faults for consideration?

Regarding Python 3.x support and a possible Biopython 2.x see
also Guido's blog post (there is probably an email version on one
of the python mailing lists too):
http://www.artima.com/weblogs/viewpost.jsp?thread=227041

Peter



More information about the Biopython-dev mailing list