Supplemental Links

An additional batch of links for your reading pleasure.

Matt Sydes at BMC blogs on sharing clinical trial data.

Brain myths die hard“. Christian Jarrett‘s reflections, as he leaves his blog at Wired.

Speaking of dying hard, Tim Lahey at The Atlantic on how medical progress is making it difficult to declare someone dead.

Money money money… where was I? Oh, yes: Dynamic Ecology on managing lab finances.

To (really) close this week’s list of links, perennial favorite Ian Street at Quiet Branches on the junk DNA brouhaha.

Share on FacebookShare on LinkedInShare on Google+Tweet about this on TwitterShare on Reddit

Monday Morning Smörgåsbord

Indira Raman has some light-hearted fun at eLife imagining Hamlet as a grant application. “Understanding the human condition is potentially of high impact. It should be pointed out, however, that previous investigators have tried and failed, so this should be seen as a high-risk-high-gain endeavor.

Much more seriously, Marina Warner at The London Review of Books on the importance of resisting the tyranny of the idiocracy. “I believe education at every age and level is an unqualified good, unassailably beneficial to the individual and to society and the world. I believe it is as important an indicator of a society’s state of health as nutrition and housing.

Another shot fired in the Statiscs Wars. Megan Hale et al in PLOS Biology propose a new approach to the problem of “(i)nflation bias, also known as “p-hacking” or “selective reporting,” is the misreporting of true effect sizes in published studies“.

Andreas Wagner at Aeon Magazine decides to poke the bear with a short stick and asks if Platonic forms can have a place in modern biology. “A systematist’s task might be daunting, but it becomes manageable if each species is distinguished by its own Platonic essence.

Redmond O’Hanlon* reviews a history of Wallace & co exploring the Amazon at The Spectator. Not a well-off Darwin in the bunch: “they were different to the norm. You could call them working-class naturalists, in that they had to pay their way by sending the specimens they captured home to their agent, to be sold to wealthy collectors. A comforting branch of English Protestantism known as natural theology had by that time become mainstream (the standard text being Archdeacon Paley’s Natural Theology, 1802). This belief that God had made everything himself (that no minor woodlouse or tiny liverwort was too small for his loving creation) led to the great aristocratic collections of the 19th century. The idea was to buy in specimens from a worker in the field — a Wallace or a Bates or a Spruce — arrange them in a mahogany cabinet, pull out the drawers, study, and thereby get to know the workings of the hand of God. This was ironic, because not one of this trio was a believer. As Wallace wrote to Bates, when they arrived in the Amazons they would study the facts of nature and work ‘towards solving the problem of the origin of species’. Which they did, eventually.

As far as I’m concerned, the take home from this Ed Yong New Yorker piece is that John Hutchinson has the funnest job in science. “Hutchinson, who is one of Kram’s protégés, planned to use similar methods for the London Zoo experiment. He is a veteran of such studies, having measured the walks of cranes, elephants, emus, giraffes, leopards, ostriches, rhinos, and tigers; his lab is outfitted with force plates in many sizes, from tiny ones for salamanders to huge ones for horses.

Writer’s block? Some tips on how to write a thesis, by Umberto Eco at the Times Higher Education. ‘Eco suggests that in writing, one should have a degree of pride: on “your specific topic, you are humanity’s functionary who speaks in the collective voice. Be humble and prudent before opening your mouth, but once you open it, be dignified and proud”‘.

Planning for the post-apocalypse? It should be you, Keith Richards and lots of roaches. If Sir Keith is too stringy for your taste, here are some insect recipes at Mosaic Science. The “dishes range from incredibly basic – simply sprinkling some crunchy bugs on top of your favourite salad – to extremely sophisticated, such as this moth mousse recipe from the Nordic Food Lab“.

* BTW if you have not read any of O’Hanlon’s books, my main reading suggestion is that you stop whatever you’re doing and go get one of them. Any one.

Share on FacebookShare on LinkedInShare on Google+Tweet about this on TwitterShare on Reddit

140 Character Problem

I’ve been both reading & participating in many interesting and enlightening Twitter discussions on science and science policy. Unlike my mostly negative experience on Facebook, the tone is almost always respectful, and when it veers towards sarcasm, it is rarely mean. Also the community structure seems to work a lot better, there is much less spam, cat videos etc… Most importantly, no one’s idiot cousin or high school friend has interjected with racist, misogynist or just plain stupid commentary. Being Brazilian in Portugal, reliably a relative or acquaintance of a FB ‘friend’ would show up making xenophobic comments.

Twitter also forces you to be concise- though no doubt in many cases at the expense of complexity and depth of argument. My most common expression of frustration is “I don’t know how to put this 140 characters!”

Most recently I internalized this sentiment (that is, I did not tweet it yet again) in a discussion with Toronto’s Jim Johnson- a discussion, I should say that I butted into, as I originally caught it when @JimJohnsonSci jumped into a thread started by ‏@pjacock (I’m still working on what is the proper etiquette to attribute Tweet credit in other media). The topic was: should there be strict publication requirements for the conclusion of a PhD degree? I’m not going to get into the pragmatic arguments, where I would mostly agree with Dr Johnson. In the past I had long discussions on this, in my role as head of our Graduate Program. Though that chalice has since passed from me to my friend Elio Sucena, I’m going to go out on a limb and say that he would agree with the broad strokes of what follows below.

The most important thing is that we need to have very high standards in what we expect of our graduate students (and must offer them the corresponding support and means). As many have remarked, the PhD is, in effect, the license to practice science (the many examples of great scientists without one notwithstanding). The candidate must demonstrate sound background knowledge, the ability to identify a relevant problem, and a tractable approach to this problem, whether experimental, theoretical or both. The candidate needs to execute this research program and arrive somewhere, presenting something novel to his/her peers.

All of this we can, and should, demand of our candidates. These things they are expected to master, control, perform. Sure, the unexpected will happen. The infection in the mouse colony. The -80 that thaws. Getting scooped. Receiving the wrong plasmid, mislabeled primers, inactive enzymes. The candidates are also expected to overcome these things. We should demand of them that they buck up and carry on. In one memorable occasion when I was a grad student, an airport shipping container flattened cages of mice we had waited some time to receive. Too bad (for the Jax Lab immigrants in particular, of course).

Publication is a different animal entirely. Too much is out of the student’s control. I’ve seen excellent papers take 2 years to get published, for no reason other than a pigheaded referee, bad journal organization, and requests for stupid, timely and laborious additional experiments. This should not delay anyone’s life, thesis defense, postdoc. Too many cases these days produce delays that in no way reflect on the quality of the candidate’s work (or the quality of the candidate). The thesis committee/jury may, more reasonably, require that a submission-ready manuscript be produced*. This too we can expect of the candidate. Publication itself, that is out of the candidate’s control, and they should not be expected to await in transit while editors and referees do their thing- particularly considering that many (most?) papers are submitted to more than one journal before finding their final resting place. This measures no quality or output of the candidate, and thus surely can not be a rigid criterium for degree completion.

*Though I would argue that here too there are important exceptions, with risky projects or truly new approaches that led to negative results, but with sterling scientific practices and sound conclusions. Strict publication requirements encourage low risk, unoriginal research.

Share on FacebookShare on LinkedInShare on Google+Tweet about this on TwitterShare on Reddit

Rendering unto Caesar

Mark Johnston has launched an intense discussion in science social media and blogs with an editorial in Genetics entitled “A Glaring Paradox”.  The crux of the argument is this: the highest impact journals (i.e. Science, Nature, Cell, etc) employ professional editors, who are no longer working in research; many lower impact journals are edited by active scientists, often leaders in their field. As Bristol University’s David Stephens put it on Twitter: “As (an) author, I at least appreciate that first cut (is) being made by (an) active scientist struggling with similar issues for publishing. Grants. Careers etc”(@David_S_Bristol). Just to be clear, what is being discussed here is the role of the editor(s), not the peer review system. The editor or editorial board makes the key decision of whether or not a manuscript goes out for review or receives the quickest (and usually least appealable) of death sentences, the editorial rejection. It is not uncommon for 80% or more of submissions to be rejected at this level.

Given this power over what represents 2, 3 or more years of hard work, it is intuitively appealing that the editorial decision be made by someone who shares the trials and tribulations of the author. But there are other considerations. Most importantly for practical reasons, the volume of submissions at some journals is not compatible with performing other professional activities. An editor at many journals will have to examine half a dozen or more manuscripts a day. One could argue that a 5 minute glance per submission by a leading specialist beats one hour close scrutiny by a professional editor who left the bench six years ago- but I would still seriously doubt that most working scientists could find those 30 minutes every day. A separate problem is that having an editor with skin in the game means that, well, the editor has skin in the game. For the platonic ideal of a scientist, this should not matter. But many of the most glaringly unfair problems I’ve seen in manuscript reviewing involved working scientists who reject manuscripts that contradict their own models. Finally, I’m not so sure that what makes a great scientist is necessarily what makes a great editor. This is similar to the relationship between artist and art critic. In some ways, and it can be very productive, as long as the critic (or editor) never loses sight of the fact that what matters is the art (or the science).

My own opinion is that much of this discussion is aiming at the wrong target. The problem is not the editorial decision at any individual journal. I firmly believe there is room for a plurality of models- even the older Proceedings of the National Academy of Sciences (USA) system of assigning publication slots to academicians was fine by me. It was the Academy’s journal, and the person who made the decision was clearly named, in print, on page one of each article. Some very good journals are run by societies or universities. Others are professional operations. The real problem, and the source of all the real anger in this debate, are the institutional decisions taken by both funders and employers in science that shortcut evaluation processes by simply counting publications in a small number of high impact journals. Science and Nature editors are currently not just making decisions on what gets published in their magazines. They are effectively the grant referees for the Fredonian Institutes of Health or the faculty search committee for Tattooine University. This is just wrong. Whatever else it may mean, as Dennis Eckmeier tweets, “Journal Impact Factor is for journals. Its value for single article evaluation is already flawed. For evaluation of a single author, useless.” (@DennisEckmeier).

If you are taking decisions about the professional future of a scientist using metrics, then we have no use for you. We can invest all your salary in a better Excel package. The right way to evaluate faculty candidates is to get to know their work. There is no shortcut. It is hard work. As EMBO Director Maria Leptin recently said, “It is the most important job you have as Director”- and that sentiment should hold all the way down the line, for anyone involved in faculty selection process. Glamour mag editors have much to atone for, but the blame for the highjacking of scientific evaluation processes by impact factor accounting does not lie with them (even if they have gleefully embraced it).

Incidentally, in my view, Science and Nature in particular should have professional editors. Because if they return to what should be their core mission, that is of publishing research that is interesting across many fields, then the editor should not be a specialist- for the  research article itself, and for the assignment (and editing, lest we forget that editors also edit) to a physicist capable of writing two pages on advances in cosmology that I, a mere biologist, can understand. Cell is a bit trickier, as it began as a specialist journal for the budding field of molecular biology. That molecular biology then did a reverse-Saturn and devoured its parent disciplines of biochemistry, genetics and cell biology effectively turned Cell into a semi-generalist journal. The ample space allowed per article in Cell, compared to the increasingly Dadaist reductionism of Nature & Science is probably also in large part responsible for the journal’s ever growing popularity.

The most worrisome segment in Johnston’s op-ed is the proposal to substitute Journal Impact Factor (JIF) with Journal Authority Factor (JAF). To be fair, it is almost certainly tongue-in-cheek. Even so, it adds to one bad problem, the innappropriate use of metrics, another, which is an implicit endorsement of arguments from authority, which are, of course, anathema in science. There is room for both professional and working-scientist (and thus, part time) editors. Each journal should decide, according to its own goals. Journal policy in this respect (as in others) should be as clear as possible, and authors can then choose the best place to submit their work. Funding agencies and hiring committees should do their own damn jobs, so that editors, professional or academic, can do theirs.

Share on FacebookShare on LinkedInShare on Google+Tweet about this on TwitterShare on Reddit

Visitors: Fred Nijhout

Friday’s seminar kicked off with a Donnie Darko moment, as Duke University’s Fred Nijhout introduced his work with a photo (sadly not reproduced here) of himself struggling to hold on to a calf-sized, somewhat menacing rabbit. Nijhout’s work centers on one of those deceptively simple questions: how do you know when to stop growing? Freaky rabbit illustrated a key point: size determination is both highly plastic and very responsive to selection. So why don’t we run into giants & dwarves all over the place? Or, as he put it, “mice and elephants start out the same size, and they never make a mistake”.

Nijhout and his group have described a variety of control mechanisms in insects. The milkweed bug stops growing when it is literally too full, responding to signals from abdominal stretch receptors. The ever-popular dung beetle is limited by how much crap their parents accumulate- the amount of fecal matter rolled up into balls to feed the larvae determines how big they’ll be. The problem is more tractable in insects because their exoskeleton limits growth to discrete periods. It also means that developing insects must determine when they are ready for a new stage, in molting or at metamorphosis.

The diversity of experimental models and approaches at one point seemed to be heading out of this world, as Nijhout setup the hypothesis that gravity was a key factor in determining size. An informal poll of the audience clearly indicated that I was not alone in expecting bugs in space. Maybe the space shuttle science payload was overbooked. The solution? ” I built a centrifuge. At 4 Gs they (the larvae) were just plastered against the side of the cage.” The solution in this particular case seem to lay in the ability to deliver oxygen to the growing Manduca.

A great seminar complete with inspirational quote. Remember kids, you’re “better off being small animal than a dead animal”.

Share on FacebookShare on LinkedInShare on Google+Tweet about this on TwitterShare on Reddit