I'm inclined to agree with the ENCODE scientists. You could argue that even protein binding sites that have no biological outcome titrate away transcription factors from other "active" regions. The system would have evolved with these sinks in place and therefore they constitute an active, useful part of the genome. Another way to look at it is that if you were trying to generate a computer model of the entire genome, you would have to account for the extra binding sites, even if they had no output in terms of protein or even RNA.
Bravo to Ars for publishing this valuable corrective, but I'm afraid that the damage has already been done. ENCODE was last week's news. People have assimilated the soundbite about 80% of the genome being functional (which probably seems like common sense to those with little or no knowledge of our genetic history and the processes that are known to shape the genome), and have now moved on.
The fact that their definition of 'functional' was utterly preposterous is a detail that will be overlooked, along with the rest of the work the consortium carried out.
The best suggestion I've heard is for the ENCODE scientists to produce a few hundred Megabases of random DNA, then test this to see how much would be 'functional' by their definition (my prediction - lots of it). Then we'd have a useful negative control and baseline.
2 comments
[ 3.0 ms ] story [ 12.6 ms ] threadThe fact that their definition of 'functional' was utterly preposterous is a detail that will be overlooked, along with the rest of the work the consortium carried out.
The best suggestion I've heard is for the ENCODE scientists to produce a few hundred Megabases of random DNA, then test this to see how much would be 'functional' by their definition (my prediction - lots of it). Then we'd have a useful negative control and baseline.