Ten years of coding with the snake

By Iddo on February 16th, 2009

Biopython is entering its 10th year; the unofficial birthday is on September, since that is when the mailing list started: September 1999. I stumbled onto that list mid-September, 1999. I believe the Python version was 1.5, I was still working on SGI Irix, and I was an 0.3 PhD candidate. Today I am coding with Python 2.5 on Ubuntu Linux 8.10, Biopython 1.49, and I am a scientist 1.9 (I hope to roll a new release of myself soon). Biopython started out when Perl dominated the Bioinformatics scripting language scene. Jeff Chang, then a PhD candidate in Russ Altman’s lab in Stanford thought it would be a cool idea to have a Python package for Bioinformatics. If there are three people that deserve credit for Biopython those are Jeff Chang, Andrew Dalke and Brad Chapman. These three contributed most of the code and design that made Biopython such a robust and useful package; of course, there were and are a lot of others involved, including myself. By the time I was finishing up my dissertation in late 2002, I think that 95% of my code was importing biopython modules, some of them coded by yours truly.

If you expected a Perl v. Python feisty language war post, forget it. The reason I started off with Python as my research scripting language is that because everybody — including Perl coders — told me Python had a gentler learning curve, and had a more “maintenance friendly” syntax. I was warned that Python was younger which means less 3rd party modules, and slower. That Bioperl was comprehensive, and Biopython was still rickety and not a lot of code there. True. But code was written, and written fast and well. Jeff, Brad & Andrew combined coding and management skills to crystallize a fantastic development community. I swear that in 10 years I have not seen one ill-willed or snarky email on either the community or the developer’s lists.

Biopython has a a vibrant, active development community.Today, Michiel de Hoon, Peter Cock, Leighton Pritchard, Tiago Antao, Brad Chapman (welcome back, dude!) , Bruce Southey and many, many others are now working hard to maintain and develop Biopython. A special hat-tip to Thomas Hamelryck who has added a kick-butt structural biology module, withstanding even the worse that PDB has to offer, something that is close to my heart. Biopython is packaged with major Linux distributions, and is part of some other bioinformatics packages. There are 150-250 posts per month on each of the mailing lists. Any way you look at it, Biopython is an open-source success story. I’m proud to have contributed modestly to this project, and I hope I will still be able to. So fire up your Python shell and type: