Date: Thu, 27 Jul 2000 12:28:55 +0200 From: Johann Visagie <wjv@cityip.co.za> To: chat@freebsd.org, advocacy@freebsd.org Cc: janet@sanbi.ac.za Subject: Bioinformatics Open Source Conference, 17-18 Aug, San Diego Message-ID: <20000727122855.B414@fling.sanbi.ac.za>
next in thread | raw e-mail | index | archive | help
Damn, I should've posted this earlier! The Bioinformatics Open Source Conference (BOSC - previously BioPerl) takes place in San Diego on 17-18 August. Unfortunately registration has already closed - hence my lament that I forgot to post earlier! See: http://ismb00.sdsc.edu/bosc2000/ I will be attending with others from my company and our associated academic institute (http://www.sanbi.ac.za/). Some background as to why this is something FreeBSD users might want to take note of (long - you may skip to summary at the end): You've probably all read the media reports about the completion of a first draft sequence of the human genome last month: http://www.sanger.ac.uk/HGP/draft2000/ Contrary to what the press hullabaloo might have suggested, this wasn't an end - it was a beginning. We haven't found the "source code" for the human species - we found the binary image in memory. Our task over the next few years (decades?) will be to disassemble that binary and get back to fully commented source code in a high level language. Calling this "difficult" may be the understatement of the new millennium. <hype> The science that is attempting this task is known as Bioinformatics. It is a already the focus of dozens of elite academic establishments around the world, and forms the backbone of a nascent multi-billion dollar industry. It is where computer science meets the biosciences - in fact, bioinformatics research requires massive amounts of processing power, and some of the bigger supercomputers in the world are dedicated to it. In terms of human scientific progress, this may well be "where it's at" for the next decade or so. </hype> Coming to this field from a commercial internet environment that would be familiar to many of you, where I too constantly fought the good fight against the hordes of Microsoft, I was amazed and not a little flabbergasted to find that here, Unix is the _only_ game in town. Literally. A bioinformatics application that runs on Windows? Never heard of one. You'd be ostracised for even suggesting it. Most bioinformatics applications are huge, monolithic, archaic, incredibly expensive and run on equally expensive commercial Unix boxes. SGI, Sun and Compaq all play big roles on the hardware side of things. They support the academic institutes with hardware, which means research crystallises into software that runs on their operating systems, which means they can move truckloads of boxes to big pharmaceuticals. As you might guess, a few insidious individuals down in the trenches have been saying, "Hey, this sucks. Look at those internet blokes developing all their software collaboratively and giving it away. We can do the same." And so a number of open source initiatives in the field of bioinformatics have been established, though most of them are still very young and immature. You can get some feel for the role of open source in bioinformatics by reading Lincoln Stein's article about "How Perl saved the Human Genome Project" here: http://www.bioperl.org/GetStarted/tpj_ls_bio.html Or see Ewan Birney's plea to help "Hack the Genome" on advogato: http://advogato.org/article/131.html Some of these projects - all of which will be the subjects of presentations at BOSC2000 - are: The Bio* projects - Bioperl, BioPython, BioJava, BioXML, BioCorba - they all have .org domains; you can guess the URLs. These tools all aim at providing the basic building blocks for developing bioinformatics applications in their respective environments. And there's a lot of cross-pollination going on too. Ensembl is a fully open source gene annotation pipeline. That'll take too long to explain, but you can find out a lot more on their site: http://www.ensembl.org/ DAS, the distributed annotation system, may be one of the killer apps of the field. You can find out more here... http://www.genetics.wustl.edu/eddy/people/robin/das And read the interesting Wired article about how DAS means gene research meets Napster, Gnutella and Freenet here: http://www.wired.com/news/technology/0,1282,35404,00.html?tw=wn20000405 Oh, and Tim O'Reilly will provide a token presence from the open source side of things by delivering one of the keynotes. Right, this is turning into an article, but let me continue. Where does FreeBSD fit in? Unfortunately, nowhere yet. As more and more effort is put into open source bioinformatics, the word on everybody's lips is, of course, "Linux". Since this is an area where massive amounts of processing power is the norm, another word that has been bandied about is "Beowulf". However, the field is rooted in commercial Unix, and you can get a feel for some of the Linux vs. Unix debate from these pages: http://www.portlandpress.com/biochemist/cyber/0006/default.htm http://www2.mrc-lmb.cam.ac.uk/personal/rpg/CB/Linux.html They detail a specific lab's decision-making process. Eventually the lab went for an Alphaserver over a couple of Linux boxes, _mostly_ because of Linux's 2GB file size limit. Erk. As a FreeBSD user in bioinformatics, I see two obvious tasks ahead of me (when I can find the time between the myriad other things on my plate): - Educating the FreeBSD community about bioinformatics. - Educating the bioinformatics community about FreeBSD. In summary: Bioinformatics is a 100% Unix-dominated field - one of the few remaining. It is about to become a Very Big Thing indeed. Unless I (and others in my position) do some advocacy of FreeBSD as a stable, secure, standard platform, it is likely to be a Linux-dominated field a few years from now. I've started working on ports of some bioinformatics tools and will submit them in due course. I'd also like to provide some feedback after BOSC if anyone is interested. (Maybe I could get Daemon News interested?) -- Johann To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-advocacy" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20000727122855.B414>