More Recent Comments

Monday, February 04, 2019

What is the dominant view of junk DNA?

I think that about 90% of our genome is junk and I know lots of other scientists who feel the same way. I'm pretty sure that this view is not shared by the majority of scientists but I don't know whether they are convinced that most of our genome is functional or whether they just think the question is unanswerable at the present time. I suspect that the latter view is more common but I'd like to hear your opinion.

How much of our genome is junk? The three most obvious answers are:
  1. Most of our genome is junk
  2. Very little of our genome is junk
  3. We don't know enough to answer the question
I recently bought the latest edition of Introduction to Genomics by Arthur M. Lesk to see how he treats genome composition. This is the third edition of a book that was first published in 2007 and it's a excellent book for learning about the common techniques in the field of genomics. However, it's not so good when it comes to discussing the big picture, especially with respect to the human genome. There's no discussion about the overall composition of our genome and how much of it is junk and no mention of the evidence supporting junk DNA. Nevertheless, the position of the author seems clear from the following passages. He is a supporter of option #3 and I suspect that's the majority opinion these days.
Approximately 5% of the dog genome constitutes functional elements common to dog, human, and mouse. This is higher than the protein-coding fraction of the human genome. It includes regulatory elements and non-protein-coding RNAs, and further suggests that it is premature to dismiss as 'junk' the regions of the genome to which we cannot yet assign function. (p. 225)
Five Things You Should Know if You Want to Participate in the Junk DNA DebateThe fact that we can assign function to more than protein-coding regions is hardly a new idea so I don't know why he makes such a big deal of this fact. Furthermore, the idea that we dismiss as junk any DNA to which we cannot assign function is a serious misrepresentation of the junk DNA position. I think it reflects a general lack of understanding of the evidence for junk DNA [Required reading for the junk DNA debate].
Regions of the genome without known function are often referred to as 'junk DNA'. Of course, the fact that we may not know the function of much of our genome does not mean that it has none. (Maybe it is junk, but it is certainly not transcriptionally inert. A series of recent discoveries has revealed many new types of RNA molecules, mostly involved in control processes. It would be naïve to doubt that many more types will come to light.) Moreover, the amount of space between genes affects the rate of crossing over and recombination and, thereby, rates of evolution. Indeed, the large amount of repetitive sequence between our genes enhances recombination rates by promoting homologous recombination. Rate of evolutionary change is characteristic of a species that is certainly subject to selective pressure. Features of the genome that affect rate of evolution cannot be dismissed entirely as junk. (p. 239)
I bet Joe Felsenstein will have something to say about selection for high rates of evolution by inserting spacers between genes. He can probably think of one or two faster, more efficient, ways of enhancing recombination if that were necessary to speed up evolution.
Traditionally, DNA outside protein-coding regions was considered 'junk', but for no reason other than ignorance. ... There has been extensive and heated debate over the classification of transcripts as functional without being able to assign specific functions to them. Part of the problem is that many transcripts for which unspecified function was claimed do not show conservation among mammals. This was the basis for many criticisms of the claims of the ENCODE consortium in its early publications. The rule, which certainly applies to proteins, that nature conserves important things but not unimportant things, does not necessarily apply to these transcripts. Sequence-structure-function relationships can be very different for RNAs and for proteins. The argument that transcribed regions are functional—rather than just 'junk RNA'—rests on the observation of differential transcription in cell-specific patterns, and dynamic regulation of transcription during development and tissue differentiation and disease. (p. 359)
As we have discussed many times, the idea that different transcripts are expressed in different cells is perfectly consistent with spurious transcription due to inappropriate binding of tissue-specific transcription factors. It is not evidence of function. Furthermore, lack of conservation is powerful evidence against function and it should not be dismissed lightly.

I think these passages reveal the true fall-back position after the ENCODE fiasco. Scientists like Arthur Lesk are reluctant to adopt the position that most of our genome is functional because of the controversy but they reject the idea that most of our genome is junk because they think it's based on ignorance. They assume an intermediate position where they claim that that we just don't know enough about our genome to decided whether it is full of functional elements or not.1

I believe that this position will not stand up to close scrutiny but that's not the point. My point is that I think this is the dominant view of junk DNA. What do you think?


1. I suspect that almost everyone who argues for option #3 is a closet functionalist.

15 comments :

  1. I can't speak for most scientists, whatever you mean by that. But I think I can speak for the small segment of scientists who are phylogeneticists who have ever worked with introns or other presumed junk. The reason we use them is precisely because they're junk, i.e. evolving neutrally. Thus the evolutionary model needed to deal with them is simpler than it might be for sequences under selection, and they accumulate changes at a higher rate and thus produce more informative sites than sequences under purifying selection. Junk: can't beat it.

    ReplyDelete
  2. Your guess as to what I would say is sound, so I won't bother saying it. Instead let me try to answer your main question as to what most scientists would say.

    Molecular evolutionists are solidly on your side (except for the usual 1% of crackpot contrarians that one finds in any field).

    Most other molecular biologists and genomicists are confused, have heard of ENCODE, know that we don't have a detailed knowledge of what is the function of everything. They may be tempted to consider the ENCODE 2012 announcement the last word. I *hope* that they have heard of the subsequent controversy. If they have they are, as you guess, likely to back off to the position that until we know all possible functions of all possible sequences, the matter is undecided (your position #3).

    It comes down to most people who are not molecular evolutionists failing to understand how strong is the indirect inference from (a) mutational load, (b) the Onion Test, (c) conservation, and (d) annotation of much of the genome as remnants of transposons and other selfish elements. They seem to think that we must await the finding of all function.

    ReplyDelete
  3. I am an ecologist with strong evolutionary interests, and throw my lot with you, John, and Joe: Junk DNA predominates in the eukaryote nuclear genome. I find the mutational load argument for non-coding DNA especially convincing. However, this does not mean (and I know these are fighting words) that I necessarily agree with points of view defended by Stephen Jay Gould. Also, I find no contradiction in considering myself an adaptationist, given proper caveats such as vestigial structures, serendipitous benefits, and the need for proper evaluation of untested adaptive hypotheses based on natural (i.e., individual), kin, group, and sexual selection (which Darwin and other ecologists treat as a separate process because it does not produce authentic “adaptations”!). Molecular drift and junk DNA seem mostly irrelevant to the evolution of whole organism phenotypes found in nature. And Joe, I still think Darwin's view, admittedly in a passing comment, that polymorphic variation may persist because it is neutral got about as close to the concept of genetic drift as possible without a knowledge of Mendelian genetics.

    ReplyDelete
  4. Haven't you recently, at least informally, acknowledged that human genome junk DNA could be as low as 60%? Correct me if I'm wrong, please!

    ReplyDelete
  5. "I think that about 90% of our genome is junk and I know lots of other scientists who feel the same way."

    Would you agree that the number of scientists supporting this view has shrunk recently?

    ReplyDelete
    Replies
    1. Your friend and ally on Junk DNA, Dan Graur, has recently adjusted his views on junk DNA.

      https://www.newscientist.com/article/2140926-at-least-75-per-cent-of-our-dna-really-is-useless-junk-after-all/

      He now thinks that ENCODE was only 60% wrong which makes evolution, as he has seen it, wrong.

      I think you should employ Dan Graur, Joe Felseintein, John Harshman and Georgio Marinov (from ENCODE) to co-write your book...

      It will kill ID all the way to the root. Finnaly!

      I could possibly help you to publish your book and market it…with the help of your committed to junk DNA friends, of course...

      How do you like it, Larry?

      Delete
    2. He now thinks that ENCODE was only 60% wrong
      I think you are about 80% wrong in reading that article. Perhaps you don't know what "at least" means?

      Delete
    3. By using genomic data from 12 mammalian species and an estimation model that takes into account functional element turnover, Rands et al. (2014) estimated that 8.2% of the human genome is functional, with a 95% confidence interval of 7.1-9.2%. Because of the difficulties in estimating the functional fraction of the genome, evolutionary biologists treat such estimates as somewhat underestimated. Thus, a claim that 10% or even 15% of the human genome is functional would be tolerable. On the other hand, a claim that 80% of the human genome is functional (e.g. ENCODE Project Consortium 2012) is misleading in the extreme and logically risible.

      Dan Graur (2016) Molecular and Genome Evolution p. 503

      Delete
    4. Jass, you didn't bother to check the journal reference at the bottom of the article, did you?

      From table 1, we see that even for unrealistically low estimates of deleterious mutation rates, the fraction of the genome that can be functional cannot exceed 25%. If the fraction of deleterious mutations out of all mutations in functional regions is even slightly >4%, then the fraction of the genome that can be functional becomes much lower. Realistically, the functional fraction of the genome cannot exceed 10–15%. These results agree with empirical estimates in the literature on the fraction of the human genome that is evolutionarily constrained (Rands et al. 2014).

      Graur 2017

      Emphasis added, in case you have problems spotting the relevant bits in a whole paragraph of text.

      Delete
    5. Gosiorowski
      How is Larry going write a book on the assumption that the functional fraction of the human genome cannot exceed 10%, meaning that 90% must be junk-DNA, if you have highlighted Graur's estimates that allow up 25% genome functionality?

      Can you see a problem here, possibly?

      Delete
    6. The majority of comparative genomic studies suggest that less the 15% of the genome is functional according to the evolutionary conservation criterion with the most comprehensive study to date suggesting a value 5%. Ward and Kellis (2012) confirmed that approximately 5% of the genome is interspecifically conserved, and by using intraspecific variation, found evidence of lineage-specific constraint suggesting that an additional 4% of the human genome is under selection (i.e. functional), bringing the total fraction of the genome that is certain to be functional to approximately 9%. The journal Science used this value to proclaim "No More Junk DNA" (Hurtly, 2012), thus, in effect rounding up 9% to 100%.

      Graur et al. (2013) p. 581

      Delete
    7. From table 1, we see that even for unrealistically low estimates of deleterious mutation rates, the fraction of the genome that can be functional cannot exceed 25%. If the fraction of deleterious mutations out of all mutations in functional regions is even slightly >4%, then the fraction of the genome that can be functional becomes much lower. Realistically, the functional fraction of the genome cannot exceed 10-15%. These results agree with empirical estimates in the literature on the fraction of the human genome that is evolutionarily constrained (Rands et al., 2014).

      Gruar, D. (2017) "An Upper Limit on the Fractional Fraction of the Human Genome." p. 1882

      Delete
    8. @Jass

      I see a problem here and it's you.

      If you can't contribute something intelligent to this discussion then don't contribute at all.

      Delete