Friday, September 07, 2012

Oh Dear. Another Non-Scientist Gets the Wrong Message from Ed Yong

David Ropeik identifies himself as an "international consultant in risk perception and risk communication, and an Instructor in the Environmental Management Program at the Harvard University Extension School." His blog is soapbox science on Nature Blogs.

Here's what part of what he posted today [A lesson from ENCODE about the limits on Human Reason].
In what should be another blow to the hubris of human intellect, we have a new entry in the long and ever growing list of “Really Big Things Scientists Believed” that turned out be wrong. This one is about DNA, that magical strand of just four amino acids, Adenine paired with Thymine, Cytosine paired with Guanine, millions of those A-T and C-G pairs linked together in various combinations to make the genes that spit out the blueprints for the proteins that make us. Or so science believed.

The problem was that, the ‘genes’ sections of DNA that coded for proteins only came to about 1.5% of the whole 2 meter-long strand. For decades molecular biologists didn’t know what the rest of the DNA…as in, nearly all of it…does. So, in a remarkable stroke of intellectual arrogance, they dismissed it as ‘junk’. Actually, the drier academics simply called it ‘non-coding DNA’. A Japanese scientist named Susumu Ohno called it junk, and the word stuck because, basically, scientists had no explanation for what most of DNA was for. So they assumed it was left over from evolution, had no current function, and was, literally, junk. As Francis Crick, one of the Nobel Prize winners for helping discover the structure of DNA, put it, non-coding DNA has “little specificity and conveys little or no selective advantage to the organism”. Right. As though nature would waste that much energy.

Well, there’s going to be a lot of editing on Wikipedia in the days and weeks to come, and it’s time to reprint the basic biology textbooks, because extensive research into the mystery of what most of DNA is doing there has discovered that the ‘junk’ isn’t junk at all. Most of it has all sorts of jobs. Science Journalist Ed Yong has written a wonderful summary of this work here.
As I said earlier, this is making my life very complicated. It's going to take a lot of effort to undo the damage caused by the ENCODE scientists and the science writers who fell for their scam.


33 comments:

  1. I asked on Ed's site, I'll ask here: Were it true that the human genome was 1.5% coding and 80% other functions such as docking molecules, then were are those other functions carried out in those bat and bird genomes that are closer to 100% coding?

    ReplyDelete
    Replies
    1. Bats and birds don't need to regulate their genes because they are just so much simpler than humans. Onions, on the other hand...

      Delete
  2. "This one is about DNA, that magical strand of just four amino acids"

    Amino acids? Looks like Ropeik has just rewritten the textbooks.

    ReplyDelete
    Replies
    1. I had to go check to make sure that was really in the original. I'd hoped that Larry had somehow pasted two sentences together and cut out a transition to a discussion of proteins. I didn't have the energy to register at Nature to point this out in the comments there. Sigh.

      Delete
    2. It was an iroic mistake, paragraph afterwards says when scientists get it wrong. They have added a clause in now to explain.

      I guess humor was lost in the midst of all this

      Delete
  3. So, in a remarkable stroke of intellectual arrogance, they dismissed it as ‘junk’.

    I seem to recall that Susumu Ohno and allies had difficulty winning acceptance for the idea of "junk" DNA. Many molecular biologists instinctively want to think of the genome as an elaborately designed machine. (If it is, think of all the grants we can apply for to work out its details).

    ReplyDelete
    Replies
    1. Yes, that's been extensively documented from the original literature by T. Ryan Gregory.

      The transcript from the 1973 talk where Ohno introduces the idea of junk DNA, and the reactions of the scientists, is priceless. See Gregory's blog for extensive quotes and sources.

      We must also keep endlessly repeating that for Ohno, junk DNA meant PSEUDOGENES, certainly not 'non-coding DNA.'

      Delete
  4. I wrote a long, fairly detailed, angry, refutation about the whole thing-- non-coding DNA = 'junk', ENCODE, Susumu Ohno, DNA made of amino acids... I don't know if my comment will get through moderation. If it does get through moderation, he'll be demolished.

    ReplyDelete
    Replies
    1. The comments up there now are pretty devastating. DM posted a response that acknowledged them but failed to address the substance at all, simply stating:

      Please note the nature of some of the comments to this piece (on the merits of the case, thank you, rather than my dumb mistake), and the same resistance to the ENCODE work in a lot of the reaction to those findings. To be fair, skepticism…challenging ideas…is part of science. I’m talking about a more fundamental intellectual/cognitive recalcitrance, and I offer other examples to demonstrate the pattern of which I speak.

      ...which would appear to demonstrate the intellectual/cognitive pattern known as "begging the question".

      Delete
  5. He does make one potentially valid point - this could spill into Wikipedia. Has anyone checked if there are major edits or edit wars going on?

    ReplyDelete
    Replies
    1. It has started: http://en.wikipedia.org/wiki/Noncoding_DNA

      When I checked, this article had this flag at the top: "This article is outdated. Please update this article to reflect recent events or newly available information. (September 2012)" and the summary section at the top ended with "However, for the human genome reports of September 2012 showed that biochemical functions were determined for over 80% of DNA."

      That last sentence should be rephrased as "...showed that over 80% of DNA is biochemically active".

      Delete
    2. Wikipedia could have a problem with this. The rules are -only from reputable sources and -no synthesis by authors. Until a refutation is published having anything other than the conclusions from ENCODE could be difficult.

      Delete
    3. I don't see a problem - the dispute is not about _what_ ENCODE said, just about the _way_ they said it (which is misleading, and has been misinterpreted in the media). So using the actual ENCODE publications (rather than secondary articles) as sources is fine. E.g. the edit I proposed above is just a less misleading way of saying exactly what ENCODE says.

      Delete
  6. Larry,

    As you know better than I, nature uses what is at hand. Why is it awkward to suppose that once some DNA has become embedded in our genome that nature would use it for whatever purpose it can put it to?

    The DNA may not necessarily have come to be in the genome as a result of a mutation that produced a direct evolutionary advantage. Presumably its existence there was a neutral accident. However, once there, since DNA can serve as binding sites, etc. it makes sense that it could serve for those effects. I don't see how that helps the creationist argument.

    ReplyDelete
    Replies
    1. This is not about the creationist argument, and it is not awkward to suppose that nature can co-opt junk DNA - we already know that this happens. This is about whether biochemical activity implies function (for any sensible definition of "function").

      Delete
    2. Where did he claim that it has no biochemical activity? Please provide a link.

      Delete
    3. Atheistoclast has no evidence for that.
      He also refuses to define "useful". Like every superstitious witch doctor, he weasels out when asked any direct question.

      Wasn't he banned last month?

      Delete
    4. Konrad, I'm not in this field -- at all. So my understanding of Larry's concern may be completely wrong. Based on your post, I'm now guessing that Larry is not unhappy with the statement that non-coding DNA supports biochemical activity. But that's a pretty bland statement. It is a chemical in a soup of chemicals. It would be unlikely to be completely inert.

      So is the question whether the biochemical activity in which it participates has any effect on gene activation, transcription, etc.

      Again, considering where it's located it would probably be hard for such biochemical activity to have no effect whatsoever.

      But if it has an effect, it is likely to be subject to selection, which would either strengthen or weaken (or eliminate) the effect.

      Given all that, would you explain what the essence of the argument is. Thanks.

      Delete
    5. That's basically right (your first two paragraphs). ENCODE claim to have demonstrated that 80% of the genome is biochemically active - which, as you point, is not a surprising finding at all. (I don't know whether anyone is challenging it, but at any rate it is not what the molecular evolution folks are unhappy about.)

      The biochemical activity is detectable, which in itself could be called an effect. The question is whether it has an effect that actually matters: i.e. is it regulatory, or does it do anything else that will affect phenotype or fitness? From the summaries I've read, the study does not seem to have addressed this question at all. This is the key point: just because something binds to something does not mean it is regulatory or that it affects phenotype. And even if it affects phenotype it may not have a detectable effect on fitness.

      I'd say to call something "functional" one has to demonstrate an effect on fitness. I'm happy to entertain other definitions, but "biochemically active" is just not an acceptable definition of "functional".

      For a longer and better explanaiton, see Michael Eisen's blog: http://www.michaeleisen.org/blog/?p=1172

      Delete
    6. Thanks, I think the follow from Eisen sums it up.

      As we and many others have now shown, molecular interactions are not rare. Transcripts, transcription factor binding sites, DNA modifications, chromatin modifications, RNA binding sites, phosphorylation sites, protein-protein interactions, etc… are everywhere. This suggests that these kind of biochemical events are easy to create – change a nucleotide here – wham, a new transcription factor binds, an splicing site is lost, a new promoter is created, a glycosylation site is eliminated.

      The neutral theory [of DNA mutation] does not demand that most sequence changes have no measurable effect on the organisms. Rather the only thing you have to assume is that the vast majority of the biochemical events that happen as a consequence of random mutations do not significantly affect organismal fitness. Given that such a large fraction of the genome is biochemically active, the same basic logic Kimura, King and Jukes used to argue for neutrality – that it is simply impossible for such a large number of molecular traits to have been driven to fixation by selection [bold added] – argues strongly that most biochemical events do not contribute significantly to fitness. Indeed, given the apparent frequency with which new molecular interactions arise, it is all but impossible that we would still exist if every new molecular event had a strong phenotypic effect.

      Delete
  7. Question for Larry? Why "wrong message from Yong"? Of all the journos that covered it, he did the best job. He did it in a nuanced way. He instantly (how many journos would do that?) came to your blog and posted a question in the comments, asking for clarifications. He keeps updating his article with all the improvements and commentary, in order to make it a lasting, useful resource. No other media outlet would ever do that. So, why bash the one journo who is behaving correctly and doing it right? You have dozens of others to name-call instead?

    I have linked to all of your posts on Twitter. I have told several people to check your posts. Because you are right. But some other people refused to tweet your links BECAUSE you are unfairly bashing the one media person who deserves to be praised, for doing journalism the modern way, who is receptive to criticism, who is not on a defensive, whose key goal is to get his article right by including commentary by experts like you.

    ReplyDelete
    Replies
    1. Hi Boris,

      Most science journalists would have recognized right away that the headline news was going to be the death of junk DNA. I think we know now that the death was greatly exaggerated.

      Knowing that this was going to be controversial, the time to contact the experts on the other side of the story was BEFORE publishing, not after. Ed should have got the other side and put it right up front near the top of the article so the he didn't leave the impression that the ENCODE result showed definitively that junk DNA was dead.

      What he's done now is to add links to other points of view way down at the bottom of the article after you've already waded through six or seven webpages.

      I like Ed Yong and he usually does very good science reporting. He blew this one. I'm sorry if some people think that criticizing a mistake by a science journalist is considered unfair. I'm picking on ED Yong precisely BECAUSE he's one of the best. I expect the bad science writers to get it wrong.

      If I were Ed Yong I'd publish a new article highlighting the controversy and apologizing for misleading so many people into thinking that the ENCODE interpretation was necessarily correct. THAT would be deserving of praise. What he's done is go part way by posting a bunch of links without distilling the science so that his readers can understand the controversy.

      The longer he waits, the less praise he will deserve.

      Delete
    2. Larry, you've known me for years. It's Bora, not Boris.

      When the embargo lifted, everyone published. I looked at many reports. Most were really bad. I kept clicking on Sandwalk and Genomicron to see you reaction - I knew they were coming. Yet, Ed Yong's was the best article, even at that initial time. He had injected multiple voices and much nuance in that discussion. Perhaps not as strongly worded as you (and I) would have liked, but lightyears better than anyone else, right there out of the starting gate.

      His post is huge. Better than anything else out there. It is getting tons of links and traffic. Any kind of separate post would not be seen by as many people, or by the same people reading the original. He explains it right there why he is updating the original post and not making a new one. And that is, IMHO, the right decision. Unlike the old-timers, Ed understands that an article is not a finished thing, then you move on. He is a modern journalist and understands that corrections and improvements must be done in the place where readers are, that an article is a living, dynamic document.

      He instantly asked you questions (has any other journo done that?). I do not see your quotes there, so I am assuming (wrongly?) that you did not provide him any - so he is a good journalist and you are a bad source. How can we expect to improve science journalism if we as scientists don't act as good sources? And yet, a big man as he is, despite you bashing him, he added links to your posts, added what he thinks are your positions. Because he respects your expertise. You should return that respect to him.

      Delete
    3. Coturnix: LM is a professor and freelance blogger. EY is a professional journalist. The tone LM used was critical but pretty unremarkable for someone reviewing work by an unconnected individual. You seem to be applying a standard of intercourse more apporpriate to one's close personal friends or collegues. I don't think universalising such a standard to all written communication is a terribly good idea--that sort of chummy approach is what makes modern American political reporting so toothless and timid, and I don't expect it to yield a much better result anywhere else.

      Delete
    4. Both Larry and Ed are friends. Both are scientists by training. Both are bloggers. Chummy is a good word to apply to blog discourse among friends, right?

      Delete
    5. Sorry Bora, I don't know why I called you Boris.

      The fact that Ed's post was huge is not admirable. You know how jouranilsm works so you know that most people don't get past the first page or so.

      You also have plenty of evidence in front of you that people who read Ed's blog came away with the impression that 80% of our genome is functional. How do you explain that since you and both know that's wrong?

      As for respect, I would have greatly respected him if he had contacted me or Ryan BEFORE he said that ENCODE has disproved junk DNA. This controversy has been around since the pilot project in 2007. That's lomg enough for the results to have been touted in Jonathan Well's book and long enough for all kinds of people to have refuted them.

      There's nothing really new in the latest results that we didn't already know.

      It's too bad Carl was busy or you would have seen how the announcement should have been handled by a science writer.

      Delete
  8. DNA, that magical strand of just four amino acids

    That just about sums up the quality of the debate.

    ReplyDelete
  9. He does make one potentially valid point - this could spill into Wikipedia. Has anyone checked if there are major edits or edit wars going on? Konrad

    Oh, dear. "Edit wars".

    Just as a point of information, but do people who could understand this issue to the point where they could have an informed scientific position on it as opposed to an ideological, uniformed opinion on it relying on Wikipedia (!) for information on it?

    Having looked at the Wiki articles on topics related to things of interest to creationism-atheism and topics related to pseudo-skepticism, I've never seen one that is reliably uninfected with dishonest, ideological "editing". That, friends, is the first thing to know about the reliability of any online source with amateur, ad hoc "editing". Making this issue a piece of turf in the war against religion is bound to make the reliability any online source of that kind justifiably suspect.

    I agree with the points Coturnix makes about Ed Yong.

    ReplyDelete
    Replies
    1. And, are people who know enough to have an informed, scientific possition on this really suggesting that those who don't go to Wikipedia to learn about it? Even as they are aware that it is a field of battle in "edit wars"?

      I remember, in high school, being warned off of print, professionally edited print encyclopedias as a source, back in the early 60s when people took the idea of objective information seriously.

      Delete
    2. Of course. I frequently use wikipedia as a starting point for learning the basics about topics I am unfamiliar with, and expect that many others do the same - that's what it's there for. Wikipedia readers generally know what they can and cannot expect from wikipedia - everyone knows it's not an in-depth treatment, and that it is edited by humans who are not typically the top experts, but the vast majority of the time it is not actively misleading.

      Certain topics on wikipedia are bound to be the subject of continuing edit wars and hence unreliable. But there is no reason why this should be the case for factual topics such as "noncoding DNA" (or most other topics in biology).

      Delete
  10. I would never use a reference source for something I knew nothing about that is susceptible to being "edited" without review (and without positive identification of the "editor) before publication. Even if, in theory, it gets corrected, eventually. One looks at a reference source at the time they look at it, not "eventually". I've read a few Wiki articles in my field that aren't bad and a few which have been excellent but I wouldn't have known that if I knew nothing about the subjects.

    ReplyDelete
  11. Faye Flam has done a good job, I think

    ReplyDelete
  12. @konrad
    You're comment is on the mark except
    "I'd say to call something "functional" one has to demonstrate an effect on fitness."
    does not seem quite appropriate to me as physical function and adaptive functionality are fundamentally different levels of inquiry(proximate- and ultimate questions respectively; See Tinbergen's four).

    Let's say for example that a genetic mutation has the effect of upregulating production of a protein involved in import of a necessary substance A and that substance A is available in a dietary abundance.
    This mutation is physically functional while adaptively neutral. The mutation does add to variation, and could easily become adaptive if dietary availability of A dwindled. So apart from being non-adaptive at the moment, variation in physical function is of significant value as a source of potential adaptation.

    Function should therefor be infered both quantitatively as qualitatively; does it produce a different result or more or less of something, it does not matter if it does so directly or indirectly.

    ReplyDelete