Tuesday, November 25, 2008

Deceptive Science?

I was recently taken in by a paper where the authors implied that they were the first to identify one of Gregor Mendel's original genes [see Identity of the Product of Mendel's Green Cotyledon Gene (Update)].

The reason I was sensitized to this issue is because of a similar incident involving a paper published in Cell. Here's an outline of the issue as reported in The Times Higher Education.

It appears that the Cell paper is very misleading. It does not give appropriate credit to the work of other labs.

Peter Lawrence asked if he could publish a letter in Cell but he was told to post comments on the website instead. This is not appropriate. Nobody reads those comments. The author of the suspect Cell paper, Jeffrey Axelrod of Stanford School of Medicine (Palo Alto, CA USA) stands by his publication. The editors of Cell use the common excuse when a journal is caught with it's pants down.
After being sent four of the letters of complaint, Professor Axelrod added on the 6 November: "Our paper (Chen et al, June 2008) underwent a strict process of peer review prior to publication. Concerns about the review process should be directed to Cell. We stand by our conclusions as stated in the paper, as well as by our use of citations."

Cell declined to respond to questions put by Times Higher Education. However, in a response to the complaint from Professor Lawrence, Cell's senior scientific editor Connie Lee said: "I can only assure you that the reviewers were experts in planar cell polarity and the consensus decision was that the model presented by Chen et al was thought-provoking, well supported and provided a sufficient conceptual advance beyond the existing literature."
The incident raises a serious question. Is the pressure to publish so great that authors are tempted to deceive editors and reviewers by making their work seem more original and novel than it really is?

This is not a new problem. A few years ago Nature was pressured into publishing an editorial defending its policies against the charge that they were publishing bad science in order to be the first journal to publish "breakthroughs" (Adam and Knight, 2002). The Nature editorial denies that the popular journals are promoting sensationalism.
Editors of leading journals reject the suggestion that standards are slipping in the face of heightened competition. "Nature has nothing to gain by the pursuit of glamour at the expense of scientific quality, considering, not least, the criticisms, corrections and retractions we would then habitually be forced to publish," argued this journal in an editorial comment earlier this month5. And many researchers who were interviewed for this article agree that the system is still working tolerably. If it ain't broke, they say, don't try to fix it.
Speaking of the pursuit of glamour, Nature continues its policy of hiring science writers to promote and hype papers that have recently been published in its journal [Human genes are multitaskers].

Adam, D. and Knight, J. (2002) Journals under pressure: Publish, and be damned... Nature 419: 772-774. [doi:10.1038/419772a]


  1. Well ... the same old bullshit. It's pervasive. Only those who profit from it deny it.

    Judging from a scandal of a couple years ago, those in power do indeed benefit from it (and that makes sense). FIVE protein structures, all published in sexy journals have been retracted:

    Supposedly, due to an bug in the program used throughout the work that resulted in the wrong handedness of the crystallographic data. The program was never named and never described anywhere along with the published data. The uproar was loud but it ended with nothing:

    1. The guy happily kept his job at Scripps (despite his tenure relying 100% on the bullshit structures).

    2. To anyone wanting to seriously pursue it, it is clear that the official explanation is bullshit as well. If there was a wrong hand in the data and they built the model incorrectly, how come the ATP binding site came out correct? (The answer is clear - it was not built based on the experimental data!)

    3. There were other, incredibly questionable technical procedures used during the work. But they got "right" results, so the dubious practices were not paid due attention by anyone during "research"/publication process.

    4. Published structures contradicted nearly all biochemical data obtained on the the proteins being investigated. That never gave a pause to any editor or reviewer in the famcy journals.

    5. Seemingly the same program and the same practices ("seemingly" because it was never described explicitely) was used by the same author to solve another sexy structure while he was a postdoc with a big shot boss. *That* structure was never retracted.

    Bottom line is that when all kinds of science-related decisions are dominated by the publication record in the big three, it is editors - not scientists! - who end up determining where and how science goes. And, because editors are normally quite clueless and make decisions based on fleeting notions of fashion and what's hot and "novel", the overall system is very counterproductive and prone to bullshit.

    It all increasingly resembles the economic bullshit bubble that just burst.

  2. The whole story stinks from the very beginning, after one asks himself why those papers about Mendel's genes even make it to Cell. Because it is not like the discovery of the positive regulator of chlorophyll degradation that's making pea cotyledons turn yellowish is of such an enormous significance for the whole scientific community. No disrespect to plant biolgists, plant molecular biology has never been paid the attention it deserves, but it is more thanobvious that the only reason why these are in Cell and Science is the history associated with them. And this is not a good scientific publishing practice. Papers with much greater and broader purely scientific significance regularly get rejected from these journals. So they compromised their standards with the very decision to publish these, no matter whether the citations were in order or not

  3. Hey Larry,

    I was wondering about your comments on the Nature articles you mention on your post (on alternative splicing).
    If you have already commented on them I apologize.

  4. Since in this case the work had been previously published, this is not the case of sensationalism winning over scientific accuracy; rather, it is the case where an unremarkable (redundant) paper gets published in a "top" journal because of political connections.
    Which happens all the freaking time.
    In fact, most papers of "top" journals are not well-cited, which confirms, most of them are indeed unremarkable.

    The real problem is when deserving papers do not get published.
    That an undeserving paper gets published, is not as bad as when a deserving paper gets rejected.

    Anyone can tell you how a large amount of truly interesting papers get rejected by top journals all the time- specially if the authors are not well-known to the reviewers, or are somehow felt to be competition or opponents of some kind by any one of the reviewers.

  5. Without meaning to detract from your main point, I'm not sure your final paragraph, poking at Nature's use of science writers per se, is an issue in the same mould as the issues raised in the rest of your piece.

    Nature (and Science) has a more "news-stand" or "general media" focus than most science journals and has long been that way. That's not a bad thing in itself: it is different, though, and in a slightly different market than most journals. I think the difference needs to be borne in mind in criticising what they do.

    Examples of its stronger media/news-stand orientation are their "News and Views" articles are written by in-house science writers (with higher science degrees I'd note); they've been around some time and I believe are aimed as much at science writers and the media as scientists. (Although admittedly probably pitched a little high for many, if not most, of them!) It runs news embargoes because the general press take material from them. And so on.

    I would like to think that their use of science writers is mainly trying to take on board some of the currently much talked-about 'science communication' issue and to provide another bridge to media complementary to those that they already do, i.e. strengthening a particular line of their business.

    There is a fine line between pursuing things too far, and maybe they haven't got it right (yet), but I don't think that their use of science writers in itself means that they are taking it too far or are in the "pursuit of glamour".

  6. Have you READ Axelrod's Cell paper or Peter Lawrence's papers to make these comments and draw conclusions? By the way, Peter Lawrence is the chief editor of the journal Development, and all of the papers he said that Axelrod had failed to cite (actually they were cited and discussed in the Cell paper) were ALL published in Development, at a surprising frequency. Shouldn't someone investigate if there is some conflict of interest or intellictual abuse there?