Ill-informed science making a case for a liberal arts education

Last month, I wrote about the terrible botch journalists had made of an interesting paper in which tweaking regulatory sequences called enhancers transgenically caused subtle shifts in the facial morphology of mice. The problem in the reporting was that the journalists insisted on calling this a discovery of a function for junk DNA — the paper itself said no such thing, but somehow that became the dominant message of the popular press coverage. Strange. How did that happen?

So Dan Graur wrote to the corresponding author to find out how the junk crept in. He found out. It’s because the author doesn’t understand the science. Axel Visel wrote back:

When I talk to general audiences (or journalists) about my research, I generally explain that the function of most of the non-coding portion of the genome was initially unclear and many people thought of it as “junk DNA”, but that it has become clear by now that many parts of the non-coding genome are functional – as we know from the combined findings of comparative genomics, epigenomic studies, and functional studies (such as the mouse knockouts in our paper).

Aargh. Non-coding is not and never has been a synonym for junk. We’ve known that significant bits of non-coding DNA are functional for a period longer than I’ve been alive…and I’m not a young guy anymore. The mouse knockouts in his paper were tiny changes in a few very short sequences — even if we had somehow been so confused that we though enhancer elements were junk, whittling away at such minuscule fragments of the genome weren’t going to appreciably increase the fraction that is labeled functional. That focus on finding more functionality in the genome flags Visel as yet another ENCODE acolyte.

Man, I’m feeling like ENCODE has led to a net increase in ignorance about biology.

Graur does not mince words in his assessment:

My problem is that junk DNA does not equal noncoding or nontranscribed DNA, and I am sort of sick to see junk DNA being buried, dismissed, rendered obsolete, eulogized, and killed twice a week. After all, your findings have no bearing on the vast majority of the genome, which as far as I am concerned is junk. Turning the genome into a well oiled efficient machine in which every last nucleotide has a function is the dream of every creationist and IDiot (intelligent designer), so the frequent killing of junk DNA serves no good purpose. Especially, since the evidence for function at present is at most 9% of the human genome. Why not call noncoding DNA noncoding DNA? After all, if a DNA segment has a function it is no junk.

Larry Moran is also a bit peeved, and explains that we actually know what a lot of that noncoding DNA does. It’s not a magic reservoir of hidden functionality.

I’ve said it many times but it bears repeating. A small percentage (about 1.4%) of our genome encodes proteins. There are many other interesting regions in our genome including …

  • ribosomal RNA genes
  • tRNA genes
  • genes for small RNAs (e.g spliceosome RNAs, P1 RNA, 7SL RNA, linc RNA etc.)
  • 5′ and 3′ UTRs in exons
  • centromeres
  • introns
  • telomeres
  • SARs (scaffold attachment regions)
  • origins of DNA replication
  • regulatory regions of DNA
  • transposons (SINES, noncoding regions of LINES, LTRs)
  • pseudogenes
  • defective transposons

These parts of noncoding DNA accounts for about 80% of the human genome. A lot of this noncoding DNA is functional (about 7% of the total genome [What’s in Your Genome?]). None of it is mysterious in any way. We’ve known about it for decades. As Dan Graur says, it’s a known known.

At least I’m in a position to do a little something about this ignorance. I’m teaching cell biology to our sophomores this semester, and next week I start the section on DNA replication, with transcription the week after. My students will know the meanings of all those terms and have a clear picture of genome organization.

And what that should tell all you employers out there is that you should hire UMM biology graduates, because they’ll actually have some knowledge of the science. Unlike certain people who seem to have no problem publishing in Science and Nature.


  1. marcoli says

    Amen. In my intro Bio class and in my Evo Bio class I talk (accurately) about the various forms of functional ‘non-coding’ DNA, and about the n-cDNA that has no known function (junk DNA). Carry on, brother! We shall over come.

  2. says

    My trouble, as a layman, is that the popular media definition of ‘junk DNA’ is the one that’s stuck in my head, not the one that scientists seem to have been using since whenever it was you guys started dissecting DNA. Maybe scientists should just go with the flow, call your ‘junk DNA’ something else, and let the media definition stand. Or come up with a snappy name for non-coding DNA (if I’m understanding the issue properly) and start pushing that in blogs and press releases.

    Meanwhile, if you want to post something explaining your list for lay-peeps, I for one would be grateful; I don’t think I have even a vague, pop-sci idea of what half of those are.

  3. Antiochus Epiphanes says

    I looked through some of the texts that I teach from, and only one indexes the term “junk DNA”; in text it appears in quotes and is in reference to the proliferation of transposons, as in “once thought to be junk DNA”.
    At least in my experience, “junk DNA” isn’t a term commonly discussed by molecular biologists because there is very little to discuss (in the absence of defending science from cretins– something many perfectly cromulent scientists spend no time doing and little time thinking about). If a biology student understands the function of regions on Graur’s list and understands that most of the eukaryotic genome cannot be classified as one of these regions, I’m not really concerned if they ever use the term “junk DNA”, the same way that I’m not concerned that my students understand baraminology.
    Related– why the eukaryotic genome contains so much more DNA than it requires is an interesting question. Early in the evolution of eukaryotes, the protonuclear genome (wherever that came from) was bombarded with cytoplasmic DNA. Extra-protonuclear DNA would have had a high likelihood of interrupting coding regions or coadapted complexes. Large regions of DNA with little function separating coding regions would have served to decrease this probability–a buffer of sorts. This paper also offers the hypothesis that meiotic recombination might have evolved in response to this kind of genomic bombardment:
    Glansdorff, N, Y. Xu, and B. Labedan. 2009. The conflict between horizontal gene transfer and the safeguard of identity: Origin of meiotic sexuality. Journal of Molecular Evolution 69: 470-480.

  4. martinhafner says

    Visel may think about junk DNA whatever he wants as long as his (IMHO wrong) opinion wouldn’t influence his data. Unfortunately, this is hardly possible because one has to make some assumptions for the extraction of signals from the raw data.

  5. anchor says

    This resonates exasperatingly with me. Oh, how many more times does one have to endure crap reporting from the Contemporary Club of Courier Journalism who (it seems appallingly evident) learned their craft licks from likes of the school of the National Enquirer…but then, like getting shot by ‘friendly fire’, one finds a primary source of such schlock schmutz: the scientists involved in the study.

    As if it wasn’t bad enough that “outreach programs” that were launched to ostensibly help public understanding of what science is, what scientists and researchers do and how they do it, and why its important to everybody, (in drooling over evident box office success of commercial enterprises) relied almost exclusively on tapping into commercial Madison Avenue talent hucksterism without ever once establishing reasonable guidelines and standards of scientific accuracy.

    Now, inevitably (as some besides me have long feared), we have out of this produced a generation of bold and ambitious researchers who have evidently been well schooled conditioned to navigate this challenging new environment that places a premium on promotional proclivity at the expense of scientific expertise. It is now okay to THINK things like, “Well, as long as it excites people it must be a positive and beneficial outcome for science”.

    The hell it is.

    Science isn’t another commodity that requires commercial gimmicks to increase its salability. Do we really think good science needs to muster a showing in a marketplace of alternatives? Are we that fucking nuts?