June 7th Auxin, microtubules, and growth, redivivus.

In which the subsquent adventures of my rant about auxin, microtubules, and growth are recounted.

Two LabFab posts this past December (here and here), covering my responses to a paper in Nature (about auxin, microtubules, growth, and thing called auxin-binding protein-1) went, not exactly viral, but perhaps bacterial. The blog got a bucket full of hits, I got e-mails from all over, and there was chatter on PubPeer, ehgo. Although a few pious souls took umbrage at my tone, which they considered flippant and personal, most were supportive. I am not going to apologize for the tone. I am a human being, it does a scientist no favors to pretend robotic detachment. My blog is personal and unfiltered, by design. I am trying, as it were, to peel the hair and scalp back from a working scientist, revealing the brain’s day-to-day goings-on. Most of the time, I write about experimental puzzles and results, intellectual stuff, fairly dull as regular readers will know, few apples bonking me, because that is what goes on. Back in December by contrast, I was angry. I saw then, and see now, no reason to have hidden that fact.

And there is news! I toned down the critique spluttered out here and sent it to Nature for consideration as a “Brief Communications Arising”, remedial comments they will post on the front web-page of a paper in question, possibly including a riposte from the authors. Nature declined it, which I was more or less expecting, after all my critique was not pointing out factual errors (e.g., the wrong number for the gravitational constant), but their decision took 2.5 months, a delay that I was not expecting. Perhaps the editors were arguing about it? Or more likely they were swamped with matters of consequence? I’ll never know.

Undaunted, I rewrote and sent it to Trends in Plant Science (TiPS). In contrast to Nature, TiPS sent the essay out for peer review, and got back three reviews, all of which were positive, in fact, breathtakingly positive. After a few tweaks (usefully suggested by the reviewers), the paper was accepted and is now in press. You can read it here  (note! This pdf has a typo – hugely embarrassing for me as well as for the journal – I hope they didn’t sack their copy editor. See if you can spot it! Sorry there is no prize other than the self-respect of a careful reader). Shout-out to Michael Wilson of CPIB who helped make the spiffy figure.

I am glad that TiPS accepted it, although sorry that their blanket prohibition on publishing as yet unpublished data prevented my including the debunking experiment. Never mind, I described it in the text and put a figure with legend and references on my web page, which the article cites. I am glad the thing is published because, of course, I love scoring points against rival scientists. As I said above about being a human being—guilty.

There is more to it than that. What follows is a rambling philosophical discourse on current trends in science (note the lower case). Broadly, I was upset about that Nature paper for two reasons. The first was its lack of skepticism. It took me two weeks to do an experiment to test the title claim of the paper. It was an easy experiment that anyone could have done. Why didn’t the authors do it? I don’t know, and in this specific case it doesn’t matter. But in general, their credulity fits the current trend of being seduced by science bling. These days, we have powerful data machines, ranging from sequencers cranking out terabytes to high-resolution confocal microscopes displaying Technicolor movies of living cells. Who wouldn’t be seduced by this beautiful information? But alas, mining data even from unspeakably powerful machines does not absolve the scientist from their duty to be skeptical and to deal with the underlying ideas as if they came from drawing lines in sand with a stick. It is easy to forget that DNA bases or pictures of cells have no intrinsic meaning.

Aligned with this, the second thing that elbowed my goat about that Nature article was the lack of scholarship. In the last years of his life, my dad was fond of exclaiming: “Convenience destroys!”. Geez dad, convenience is a good thing. Who wants to read by candlelight, draw water from a pump, or cut quills to make a pen nib? Anyway, quills, pumps, and candles are the conveniences of an earlier age. But my dad had a point, when you look, it isn’t difficult to find the destructive side to any convenience: candles burn down houses and cell phones are built in sweatshops.

As the output of science has gone from printed journals to the convenience of electronic files, it is worth asking what has been destroyed. Or at least lost. As an object held in the hand, a journal issue had gravity, not quite that of a book, but all the same an object with mass. The weight of the printed journal tended to support, even demand, strong production values, ranging from editing to design and manufacture. Now, with the convenience of merely hosting a set of files, our output seems to be loosing its integrity. Editing is often minimal or even absent and design is increasingly ignored; the task of writing relegated to hoaking up a few sentences to point at the experimental results. Matters of scholarship, like editing and design become old-fashioned and optional. The extreme exemplar of this view is the idea that the scientist’s job requires nothing more than uploading their raw data into some cloud.

I am glad that my critique was published because it shows that the emperor, in the guise of a high-profile Nature paper, short on scholarship and logic, has no clothes.