Denis Noble writes about junk DNA

I have read Dance to the Tune of Life. It's a very confusing book for several reasons. Denis Noble has a very different perspective on evolution and what evolutionary theory needs to accomplish. He thinks that life is characterized by something he calls "Biological Relativity." I don't disagree. He also thinks that evolutionary theory needs to incorporate everything that has ever happened in the history of life. That's where we part company.

I'm working slowly on a book about genomes and junk DNA so I was anxious to see how Noble deals with that subject. I tend to judge the quality of books and articles by the way they interpret the controversy over junk DNA. Here's the first mention of junk DNA from page 89. He begins by saying that it's difficult to explain development and the diversity of tissues in multicellular organisms. He continues with,

Not only is that diversity difficult to explain. The diversity amongst different species is also still difficult to explain. We humans share nearly all our genome with that of a mouse. With just a few differences, virtually all the same proteins are made using the same germline genome templates. That is even more true when comparing us with a monkey species. And even comparing humans with a simple worm, like C. elegans, we find that we have similar number of 'genes' (as protein templates), which is between 20,000 and 25,000. There are plants that have more 'genes' than we do. It could be that many of the important factors responsible for the differences lie in the regions of the genome that were once once labeled as 'junk'—meaning without function—which accounts for more than 95% of the human genome. That idea of 'junk' DNA has now itself been junked as we have found that the great majority of the sequences are transcribed into RNAs. Many RNAs have function within the organism, for example in controlling the genome, and some of them are also inherited independently of DNA. This takes us into areas of modern biological science that are at the very frontier of current research. Epigenetic control of the genome will be discussed in chapter 8.

This is a remarkable passage for several reasons. It's surprising to me that after all the discoveries of the 1980s Noble still doesn't understand the basic principles of developmental biology. I think I have a pretty good idea why whales and humans look so different in spite of the fact they have similar genes. It's because of differences in the expression of those genes during development. I don't find that difficult to explain. We don't need to look for extraordinary explanations when transcription factors and regulatory sequences—known since the 1960s—are quite capable of creating diversity.

I'm also surprised that after all the public controversy over ENCODE and junk DNA, he still thinks that pervasive transcription rules out junk DNA. If he actually believes that most of our genome makes important regulatory RNAs then that's a real problem for the definition of 'gene' that he discusses at numerous points in the book. In several passages he attacks the traditional molecular definition of "gene," which he thinks is equivalent to "protein-coding DNA."¹

The only other place where junk DNA comes up is on pages 228-229. He seems to acknowledge that there's more to evolution than traditional genes. Clearly ENCODE leaders must have known this. But that doesn't prevent Denis Noble from criticizing his whipping-boy version of gene-centric evolution.

Isn't a Lot of DNA 'Selfish', 'Parasitic'?

One of the standard defenses of selfish gene theory is based on the discovery that, in humans and many other organisms, only a few percent the genome codes for proteins and can therefore be classified as 'genes' in the original molecular biological sense. The rest was described as 'junk' DNA, the ultimate example of 'selfishness' since it was seen as DNA 'hitching a ride' with no function, a bit like a virus that has become permanently resident in the body. The strong implication is that this discovery favors the selfish gene view.

In order to understand this passage, you need to know a bit about Denis Noble's view of evolution. He thinks modern evolutionary theory is based on The Selfish Gene by Richard Dawkins. He also thinks Dawkins was referring to real genes rather than metaphorical genes. That's the same kind of misunderstanding that I discussed in covering the Lu and Bourrat paper. There seems to be a widespread belief out there—a belief shared by both philosophers and biologists—that evolutionary theory is based solely on natural selection acting on traditional genes.

Noble shares another widespread misconception concerning junk DNA. He thinks the problem of junk DNA was solved in 1980 when a series of paper appeared in Nature. Those papers tried to account for extra DNA by suggesting that it was selfish DNA consisting mostly of transposons and viral DNA. These people think that the problem of excess DNA was incorporated into, and fully explained by, selfish gene theory.

The truth is very different. Less than 1% of the excess DNA in our genome consists of functional transposons and viruses.² You can't explain junk DNA by pretending that it's selfish in the sense used by Dawkins or the authors of those Nature papers. 90% of our genome is real junk that's not "selfish" at all by any definition of the word.

Noble's view leads to confusion on many levels.

We now know that it's a confusing way of viewing genomes. There are several ways in which the confusion can be unraveled.

The words selfish junk etc. are, of course, metaphors. More importantly they are empirically empty metaphors when applied to sequences of DNA. No conceivable experiment could validate or invalidate them, as explained in chapter 5. That the metaphors are empty from the standpoint of empirical science does not mean they have no impact. On the contrary, they have had, and still have, very persuasive impact on the way in which many people think about biology.

If certain DNA sequences are junk then the best way to prove this is to remove them and see what effect they have on the organism and the population. Some of those experiments have been done. There are other ways of gathering evidence for function or for junk. It is not true that the claim of abundant junk DNA is immune to scientific investigation. That's a very strange statement.

It's true that you can think of "junk" and "selfish" as metaphors. But these words also have real meanings in some contexts. When Dawkins refers to "selfish genes" he is using both words metaphorically. He means any bit of DNA that's responsible for a phenotype that can be acted on by natural selection.

When Orgel & Crick refer to "Selfish DNA: the ultimate parasite" they are referring specifically to transposons and viruses that are "selfish" in the sense they act only for their own survival without regard to the needs of the organism. This is the same meaning used by Doolittle & Sapienza when they wrote about "Selfish genes" (Orgel and Crick, 1980; Doolittle and Sapienze, 1980). This distinction between different meanings of "selfish" and "gene" was hashed out in a series of paper back then [see, Selfish genes and transposons]. It takes a bit of work to understand what's going on but that's what you are supposed to do if you are attacking selfish genes.

A more persuasive counterargument is therefore needed. Recent experimental work has provided us with precisely that. The more we examine non-protein-coding DNA the more evidence we find that an overwhelming 80% is transcribed to form RNAs, and that around 20% of those are already known to have function. It was far too premature to dismiss the great majority of the human genome as junk.³

This is called having your cake and eating it too. First, you declare that no experiments could prove or disprove the existence of junk DNA. Then you promote an experiment that presumably does just that. The experiment confirms your bias; namely, that most of our genome is functional and that explains species diversity. Naturally, in order for this logic to work you have to ignore any experiments that support junk DNA or any criticisms of the experiment you mention. Denis Noble does that very well.

This also raises questions about what precisely is a gene. In the original sense defined by Johansson and 1919 1909 (see chapter 5) a gene is anything that contributes to inheritance. At the least, therefore, all the DNA that does that should be regarded as genes. It is too restrictive to define 'gene' as only protein-coding DNA. On a relativistic view we should, however, go even further than that. As we have seen from Waddington's and other experiments, the whole pattern of the genome is relevant to inheritance. The real problem with the gene-centered view is that it ignores the property of the genome as a whole. These overall patterns may have functional significance through the eyes of a higher-level analysis at which functional organization and processes appear.

It's amazing to me that critics of Dawkins can be so confused. When Dawkins talks about selfish "genes" he is using the word "gene" to mean exactly what Johansson meant more than one hundred years ago. Here's how Dawkins himself explains it in The Selfish Gene (p. 28).

A gene is defined as any portion of chromosomal material that potentially lasts for enough generations to serve as the unit of natural selection.

How could that be any clearer? Dawkins has no problem incorporating non-protein-coding molecular genes into selfish gene theory.

The problem with evolution critics like Denis Noble is twofold: (1) they believe that selfish gene theory is equivalent to the Modern Synthesis and that it represents the dominant view of 21st century evolutionary biologists and (2) they think the way to attack that theory is to focus on the meaning of the word "gene." In this sense, Denis Noble's thinking is not much different than that of most philosophers of biology, which leads me to believe that those philosophers are relying on popular books for their knowledge of evolutionary theory.

For the record, my view of current evolutionary theory is based on population genetics at the lowest level. This is the level of DNA and the variants that we are concerned with are alleles, not genes. Alleles can be found within genes or in other parts of the genome. This includes functional elements such as centromeres, regulatory sequences, and origins of replication that aren't "genes." It certainly includes non-protein-coding genes such as those for snRNAs, microRNAs, and tRNAs. Evolution can also occur at the lowest level by changes in allele frequencies within junk DNA. The main mechanisms of evolution are natural selection and random genetic drift.

Alleles can beneficial, detrimental, or effectively neutral. Population size is important. The selfish genes of Dawkins are just one part of evolution at the DNA level. There are many important parts of evolutionary theory that deal with higher levels such as speciation.

The current model of evolution is not gene centric as most of its opponents argue. It is based on population genetics, to be sure, but the appropriate definition of evolution is "change in allele frequency" not change in gene frequency. You could say that it's DNA centric but even that's an exaggeration because it's mostly population centric.

But the purpose of this post is not to engage in ridiculous quibbles about the meaning of evolutionary genes and molecular genes and why they aren't the same. Nor is it meant as a critique of current evolutionary theory. The purpose is to see how Denis Noble interprets the junk DNA controversy in order to get a measure of his ability to understand and explain complex issues in evolutionary biology.

I now have a good understanding of his ability.

---

1. Denis Noble first defines the molecular definition of a gene as, "a sequence of DNA that is transcribed to produce a functional product." That's remarkably similar to the definition I prefer, "A gene is a DNA sequence that is transcribed to produce a functional product" [What Is a Gene?]. Shortly after that he reverts to attacking the molecular definition because it's restricted to protein-coding regions.

2. A large percentage (>50%) our genome is junk DNA derived from pseudogenes, defective transposons, and defective viruses. These sequences were once functioning elements but they can no longer be referring to as "selfish" DNA in any meaningful way.

3. Noble inserted two references in this paragraph. The first was in support of his claim that 20% of RNAs are known to have function. The reference is to an article on ENCODE published in Science in 2004!!! The second reference supports his claim that it's to premature to conclude that most of our genome is junk. That reference is ... wait for it ... Nessa Carey's book "Junk DNA"!!! It's clear that Noble's knowledge of this topic is very superficial.

Doolittle, W.F., and Sapienza, C. (1980) Selfish genes, the phenotype paradigm and genome evolution. Nature, 284:601-603. [PDF]

Orgel, L.E., and Crick, F.H. (1980) Selfish DNA: the ultimate parasite. Nature, 284:604-607. [doi: 10.1038/284604a0]