I am currently in the process of reading Francis Collins‘ most recent book, The Language of Life — DNA and the Revolution in Personalised Medicine.

I have to confess to an element of surprise when I read the following statement on page 6 of his book:

The discoveries of the past decade, little known to most of the public, have completely overturned much of what used to be taught in high school biology. If you thought the DNA molecule comprised thousands of genes but far more “junk DNA”, think again.

On page 9 he comments on the constitution of the genome, noting that exons and introns of protein-coding genes add up together to about 30% of the genome. But here’s the astonishing thing, coming from Collins. Regarding the long segments of DNA that lie between genes that don’t code for protein, he tells us,

These regions are not just filler, however. They contain many of the signals that are needed to instruct a nearby gene about whether it should be on or off at a given developmental time in a given tissue. Furthermore, we are learning that there may be thousands of genes hanging out in these so-called deserts that don’t code for protein at all. They are copied into RNA, but those RNA molecules are never translated — instead, they serve some other important function.

Is this really the same Francis Collins who wrote The Language of God, in which he tells us that it “strains credulity” to think that more than a few pieces of “junk DNA” could be functional in the cell? Compare these remarks with his previous statement in The Language of God:

Darwin’s theory predicts that mutations that do not affect function (namely, those located in “junk DNA”) will accumulate steadily over time. Mutations in the coding regions of genes, however, are expected to be observed less frequently, since most of these will be deleterious, and only a rare such event will provide a selective advantage and be retained during the evolutionary process. That is exactly what is observed. This latter phenomenon even applies to the fine details of the coding regions of genes… If, as some might argue, these genomes were created by individual acts of creation, why would this particular feature appear?

But is this what is observed? On page 293 of his new book, Francis Collins tells us:

It turns out that only about 1.5 percent of the human genome is involved in coding for protein. But that doesn’t mean the rest is “junk DNA.” A number of exciting new discoveries about the human genome should remind us not to become complacent in our understanding of this marvelous instruction book. For instance, it has recently become clear that there is a whole family of RNA molecules that do not code for protein. These so-called non-coding RNAs are capable of carrying out a host of important functions, including modifying the efficiency by which other RNAs are translated. In addition, our understanding of how genes are regulated is undergoing dramatic revision, as the signals embedded in the DNA molecule and the proteins that bind to them are rapidly being elucidated. The complexity of this network of regulatory information is truly mind-blowing, and has given rise to a whole new branch of biomedical research, sometimes referred to as “systems biology.” [emphasis added]

I’m sure that many readers will be rather surprised to learn of Collins’ change of tune. One is, of course, tempted to wonder about the extent to which the efforts of the modern ID movement over the last decade have played in all of this, which have certainly popularised the notion that so-called “junk DNA” is not, in fact, necessarily junk after all. The evidence continues to pile in, documenting more and more numerous instances of functionality with respect to these non-protein-coding elements, even in some cases in the non-transcribed portion of the genome. Francis Collins’ darwin-of-the-gaps has certainly shrunk several fold since he wrote The Language of God in 2006!