Biology and Black Holes

I attended a talk Monday in which the speaker proposed that biology was on the cusp of a new era. This era, he claims, will see biology increasingly resembling the discipline of astronomy. We will more and more be training our genomic telescopes on small parts of the genome, building models, and testing that model against other parts of the genome.

While I don't disagree with him that this is where the field is going, I think we need to do everything we can to avoid the possible consequences. The astronomical model seems, to me, to be a black hole of correlative, descriptive analysis with no useful predictions or clear connection to human application. We risk getting lost in a biological string theory, which makes no substantive claims but which drains brainpower and dollars from more applicable efforts.

What do I mean by this? Take CHiP-seq studies, for example. These are incredibly powerful experiments that document behavior of transcription factors throughout the genome, and they reveal new interactions and possible regulatory pathways. But they often do so in a deluge, and teasing apart the specific and biologically relevant (read: predictive and useful) associations from the less specific, less relevant results can be an entire career's worth of work. In short, we have more data than we know what to do with.

I don't want to sound totally negative about biological astronomy. If we train our telescopes specifically towards disease states we may be able to sort out relationships between genomic, epigentic, trancriptional or other states and prognosis or even treatment responses. Massive biological data coupled with excellent clinical annotation could go a long way towards personalized medicine. But in the complex regulatory network of the cell it seems unlikely that any simple interpretations for these data sets will emerge anytime soon. For now, we should expect biological astrology; We can make some predictions about the future but be damned if we know anything about the mechanism.