[Biopython-dev] Biopython 1.60 plans and beyond

[Eric Talevich]

On Sat, Feb 18, 2012 at 4:39 AM, Peter Cock <p.j.a.cock at googlemail.com>wrote:

> Hi all,
>
> Assuming we ship Biopython 1.59 soon (Feb 2012),
> we should start thinking about what is ready to merge to
> the trunk afterwards to be included in Biopython 1.60,
> and what else is being worked on beyond that. This
> might also help with GSoC project ideas.
> http://lists.open-bio.org/pipermail/biopython-dev/2012-February/009383.html
>
>
My thoughts:

1. Lingering GSoC code:

(a) João has some interesting features on his own fork that are not yet
complete. I believe he's using this code for his own work and ironing out
the details, and will let us know when it's ready to merge.

(b) Mikael Trellet's Interface work is good, but not yet merged. Soem
dependencies on João's work, maybe?

(c) Michele Silva wrote two modules under Bio.PDB that use mocapy++ to do
stuff. It's based on Thomas Hamelryck's published work, with some
modifications to improve Biopython integration. I think it could be merged
cleanly at any time, under either Bio.PDB or the proposed Bio.Struct.

(d) I squashed Nick Matzke's Bio.Geography branch into a single commit and
put it on another branch on my public fork. It needs work, and some day
I'll probably take care of that.


2. Bio.Phylo improvements:

(a) ETE offered use a public-domain NeXML parser; "all" I have to do is
copy it into our codebase and convert/wrap parts of it to fit Bio.Phylo's
Tree/Clade object model. Or something along those lines.

(b) I have a RAxML wrapper that's mostly complete. It will go in
Bio.Phylo.Applications.

(c) I'd like to add functions for majority-rules consensus and
Robinson-Foulds tree distance. Then I'll consider it pretty much
feature-complete, pending feature requests from users.


3. SeqIO read-only support for PDB files (
https://redmine.open-bio.org/issues/3295). I've been using this code on my
own. It fails to parse at least one PDB file I care about (3BEG); I haven't
tried it on a larger set of PDB files. In any case this shouldn't be too
hard to fix, and I'd like to see it in a stable Biopython release.


Official releases for Python 3 (even if we do call it
> a beta status release). Maybe we can even do this
> with Biopython 1.59? Most things are working (with
> the exception of some C code and missing third
> party dependencies), and my concerns about the
> memory overhead of unicode strings should be
> resolved with Python 3.3 (the parsing speed
> overhead perhaps not).
>
>
I sense that Python 3 users are aware there can be minor performance
regressions relative to Python 2, and this applies to libraries too. (I
could be mistaken.) So I'd be in favor of making Biopython 1.59 an official
release for Python 3, but marked "beta" in the release notes since not all
parts of Biopython are fully functional in Python 3.

-Eric