[Biopython-dev] Fwd: Abstract

[Tiago Antão]

Hi all,

We are trying to submit an abstract for BOSC 2008 regarding Biopython.
Below is the current version.
Comments would be very appreciated (we are already after the deadline,
so they should come in fast ;) ).
Michiel, do you want to add anything to the "future" section?


---------------------------------------

Biopython Project Update

Tiago Antao[1], Peter Cock[2]

In this talk we present the current status of the Biopython project,
we focus on features developed since BOSC 2007, future plans for the
project and present example usages of the new population genetics
module.

The latest Biopython release is 1.45 made available on 22 March 2008.
Some of the new features are:

  1. A new population genetics module including support for
coalescent simulation, selection detection and the GenePop file
format. The new module relies on existing open source external
software (e.g., the open source Simcoal2 for coalescent simulation
which is can take advantage of multiple core CPUs for computationally
intensive tasks).
  2. Improved documentation.
  3. Deprecation of many modules which were either obsolete or had
been superseded by other code.
  4. Plus many bugs were fixed, included updates for evolving file formats.

Since the Biopython 1.45 release, further work is planned to extend
the Population Genetics module (e.g., with a statistics component).  A
new sequence alignment module is also being implemented with a uniform
API for reading and writing various alignment files, based on the
approach of the Bio.SeqIO module added last year for working with
sequences.  Work to improve Biopython's BioSQL support is also
ongoing.

Time permitting, the talk will also show usage examples of the new
population genetics module. The focus will be put not only on the
population genetics side, but also on strategies to easily use all
available computational power on new multiple core computers. This is
useful for users of the most scripting languages as most language
interpreter implementations impose stern limits on multi-threaded
programming efficiency, which is important when using computational
biology code which is CPU intensive. We will take this opportunity to
discuss strategies to overcome those language limitations.



Any feedback would really be much appreciated, thanks!