Panglossian genomics

As most French students of my generation, I had to study Candide, a short philosophical novella written by Voltaire. Back then, I was convinced that Voltaire was an arrogant prick, and I never imagined that his dumb criticism of Leibniz’s theory of pre-established harmony, which he barely understood, would ever echo in my work as a biologist.

But here we are, years have passed, I have made peace with Voltaire, and the ENCODE consortium has issued its major and controversial statement that they find “biochemical functions for 80% of the genome”. As the arguments and the comments flow on the blogs and in the academic press, I cannot help thinking about the words of Dr. Pangloss – incarnating narrow optimism.

Observe, for instance, the nose is formed for spectacles, therefore we wear spectacles. The legs are visibly designed for stockings, accordingly we wear stockings.

What I will call the Panglossian reading of the “80% functional” statement above is the idea that 80% of the genome is meant to be the way it is. The architecture of a given locus is somehow designed to produce what happens there (transcription, transcription enhancing, transcription factor binding etc). Notice how this notion of “biochemical function” is close to the fundamentals of Intelligent Design. I would not be surprised if the above statement ends up being taken over by some creationist movement – unless 80% would be considered too puny for an act of God.

Junk DNA

To understand what is behind the statement of the ENCODE consortium, we need to go back to the battle over junk DNA. As the Human Genome Project was ploughing forward, it was surprising – not to say disappointing – to realize that protein-coding sequences make up only 2% of the genome. In other words, all the constituents of your body and your metabolism (and there are a lot) represent only 2% of the genetic material you inherit from your parents. What does the remaining 98% consist of? Mostly repeated sequences such as transposons, dead viruses, by-products of their activity such as Alu repeats and introns (the non-coding parts of genes).

The term ‘junk DNA’ coined by Susumu Ohno in 1972 often designates this apparently wasteful load of non coding DNA. Over time, mass media and fiction have loaded the expression with even more emotional an irrational connotations than it was given at birth. My personal favourite is the TV series Dark Angel, where you learn that the inexplicable super powers of the hero come from the fact that she has no junk DNA (sorry for the spoil, but don’t watch it... really).

Overall, it is hard to tell whether the battle for junk DNA is ideological or scientific, but the position of the ENCODE consortium is clear: it is not junk after all!

Wegenerian genomics

I have to confess that this debate over junk DNA has always made me feel uncomfortable. Genomes are natural phenomena, not engineered devices. The whole discussion whether some parts are functional or not is a misconception and an insidious entry point for the notion of purpose (more on this in Lessons from Intelligent Design).

Ironically, best-sellers such as The Blind Watchmaker by Richard Dawkins, advocating for a universe without design have convinced the community of just the opposite, otherwise how would the debate even take place? If there is a function, there is a design. Whether the designer is God or natural selection, you end up dangerously close to the Panglossian point of view.

Is there space for non functional, or non Panglossian genomics then? I believe there is. Nobody would discuss the function of the north face of mount Everest for instance, it just happened as a consequence of continental drift, the theory proposed by Alfred Wegener. In a similar way, genomes are the product of biological evolution, a process not yet fully understood. Asking what parts of the genome are for is interesting for bio-engineering, but it makes us forget that we do not fully understand where they are from.

Natural selection is only one of the many aspects of biological evolution. Again according to the ENCODE consortium, it maintains a small fraction of the human genome (less than 10%). What other forces flesh out the bulk of the genome is still an open question.