O'Really?

Life is complicated, full of “emergent” properties

Drew, and many people like him, would like Biology to be more like Engineering. But we’re not there yet. As Pedro points out in Design, mutate and freeze, we can’t yet cut-and-paste arbitrary biological parts out of the bacterium Escherichia coli into other organisms like yeast, and just know they will work in the same way. This is because biology is complicated, and our knowledge is very incomplete. One of the main tools we have for understanding life from the molecular level upwards is reverse-engineering. You take a system and pull it to pieces. This is the reverse of conventional engineering where you start with standard and well understood parts and build them into a potentially complex system. The classic paper describing these different kinds of “engineering” is Yuri Lazebnik’s amusing article [1] which speculates how Biologists would set about understanding radios. Most readers of Bio::Blogs have probably already seen it, but if you haven’t, go and read it now!

Whereas Biology is complicated, engineering favours simplicity. Take XML for example, a standard piece of engineering which recently celebrated its 10th birthday, see Tim Bray’s XML people for a long-winded but interesting account of the social history. XML is widely adopted in many different areas of biology, with virtually no fuss or hype whatsover. Why? Because it is relatively simple, and it just works. So we already have many success stories based on XML such as the National Library of Medicine DTD for describing publication metadata in XML. We also have the Systems Biology Markup Lanauge (SBML), adopted by hundreds of software companies allowing interoperability of applications using XML. The simplicity of XML, and the quality of the tools that surround it is one of the main reasons behind these success stories. This is something the semantic web community could learn a lot from, as their favoured engineering standard (something called RDF) celebrates is 9th birthday, with a little help from Dave! Beckett! at Yahoo!. Before I go off on another semantic web rant, let’s move on. Engineers like simplicity, but Biology is complicated.

Engineering contributions to Biology

Back in the world of engineering Science Writer Philip Ball wonders about the Engineering Grand Challenges in his post Engineering for the better? As he puts it:

“It’s heartening to see engineers, long dismissed as the lumpen, dirty-handed serfs labouring at the foot of science’s lofty citadel, asserting in this manner their subject’s centrality to our future course.”

Well Hear Hear! That man talks a lot of sense, somebody should buy him a beer. There are a total of 14 grand engineering challenges listed, many with a significant biological component of some sort. Let’s pick just three examples:

1. Engineer better medicines for commentary on these see Deepak Singh on Engineering Better Medicines where he points out that tomorrow’s healthcare will be driven by a mix of chemistry, biology, informatics and engineering and Ouroboros on bioengineering approaches to understanding ageing
2. Reverse engineer the brain We’d get a better understanding of the brain, and more intelligent computers as a result. Making Artificial Intelligence work is a massive challenge, Mind Hacker Vaughan Bell has some interesting commentary on Blue Brain Rising
3. Engineer the tools of scientific discovery This is one area where bioinformatics already plays a huge role, and will continue to do so for many years to come…

On the subject of interdisciplinary informatics and engineering: Richard Apodaca was busy demystifiying Java applets which have long been a clumsy solution, unlikely to facilitate scientific discovery. At least now, they are easier to use. Thomas Lemberger wonders about the role engineered machines and databases might play in publishing and Egon Willighagen wonders the same, about JANE, for discovering interesting papers. Both of these echo some of Phil Bournes comments about the convergence of publishing and databases at How a Biological Database will be different from a biological journal?

Meanwhile, Bosco Ho demonstrates how to use simple html to visualise large chunks of data and Neil Saunders explained how to map protein sequence onto chromosomal coordinates using BioPerl while pondering the merits of workflows. Michael Barton also outlined some of the challenges of re-using code in software engineering and bioinformatics. Data integration is always a massive engineering challenge in bioinformatics projects, Rod Page was wondering How shall I integrate thee? Let me count the ways… with his LSID Tester, a tool for testing Life Science Identifier resolution services. Good engineering and great tools can empower scientists to do their work more efficiently, Cameron Neylon thinks web feeds are just the ticket and that they will help him rule the world. We’ve just started collaborating on LaBlogs, Laboratory Logs / Weblogs which will be an interesting project. Talking of blogs and laboratory notebooks, Jean Claude-Bradley was wondering about Open Science again, over at UsefulChem

The “interdiscipline” of engineering, biology and informatics is often a challenge in itself. Materials scientist (or is that engineering?) Brian Derby was wondering if interdisciplinary science is always “lopsided”, with one side providing a target or a developed technique for the other to work with. He uses the Physics / Life Sciences interface as an example, though I suspect a lot of what he says applies to Bio-Engineering also. For example, how much Bioinformatics is all Bio- and little -Informatics, or all -Informatics and not much Bio-?

Engineering or Biology?

All this interdisciplinary goodness brings us back full circle, from Biology to Engineering to Biology again in the shape of DNA computing. Biocomputer Martyn Amos, when he’s not having sex with robots on the Oxford Road in Manchester, has authored a book about DNA Computing (Genesis Machines: The New Science of Biocomputing pictured right). In his post, Biological complexity: from molecules to systems he outlines the speakers (of which he is one) for a very interesting looking shindig in London later this year.

Looks like great stuff, is it engineering, biology or both? That rather artificial distinction will get increasingly blurred, perhaps it’s meaningless already?

References

1. Yuri Lazebnik (2002) Can a biologist fix a radio? or, What I learned while studying apoptosis. Cancer Cell, 2(3):179-182. DOI:10.1016/S1535-6108(02)00133-2 (free version available from pubmed.gov/15627398)