Supplementary?
Posted by benoitbruneau, on 19 August 2010
It seems that following on the tracks of Cell Press, which is reducing the maximum number of supplemental figures to one per manuscript figure, now J. Neuroscience is doing away with it altogether. Hooray?
I agree that it is not a very good thing at times that the amount of Suppl Figs has risen (or sunk) from useful to occasionally ridiculous (20+ figures!). Yes, it is useful to be able to add a few control experiments, or the validation of a mouse knockout, and a good place to put especially large datasets, but now it’s become an excuse to either bury data that isn’t super solid (in hopes that reviewers won’t pay too much attention), or from the other side, an open invitation for reviewers to ask for more (It’s like a reverse Oliver Twist: “Can I have more please?”; “MORE!?”).
And for the most part, it’s a rather annoying exercise to have to go download the suppl materials. Why can’t journals not put the supplementary pdf together with the main paper pdf? I do it all the time, it’s a simple feature in Acrobat, so is that so much to ask for from a publisher? Cell Press does it, Nature Cell Biology does it, but that’s about it.
So, Supplementary/Supplemental figures/data? Good, bad, ugly? Discuss. Perhaps we can help influence some journals that are paying attention.
(3 votes)
Loading...
For start, one huge mistake that should be corrected soon for the main journals is this: supplemental methods should not exist. Many journals forget the main idea of a paper: an instrument to other scientist, in the most complete form in order to be independently replicated by other scientists. Today, a paper, more than an instrument, seems like a commercial product, and many scientists try to hide relevant information, and many journals try to hide the methods to give more space to ads and more papers in the printed version.
A paper should include complete information about the methods, and references included in this section should be verified by the editors (for example, the classic “Microinjections were performed as previously described in reference 7”, and then you go to reference 7 and find that you have to go now to the reference 9 in the paper cited in 7, and so on, until getting into the paper describing the method; that’s annoying).
About supplemental figures, yes, there should be a limit in the number, and precise rules about the content of them. For example, validation of siRNAs and knockouts, the set-up of a technique, and so on. Trying to demonstrate in a supplemental figure that a specific treatment, not fully discussed in the main article, has a given result just to ensure the novelty of a finding, is not enough and should be eliminated from the article.
Benoit – thanks for your interesting post on this thorny topic. We discuss what to do with supplementary data at every Development Editors meeting, and what to do about the increasing demand from reviewers for more and more data, much of which inevitably ends up as supplementary material online.
I’d be really very interested to hear the community’s thoughts on this topic so please do write in with your comments. We are listening…
BTW in response to Pablo’s comment, we don’t allow Materials and Methods to be published in the supplementary material at Development because we believe that the information required to replicate the reported experiments should be included with the findings of the paper.
+1 to Pablo’s comment (and Jane’s response) re: methods. When I read a paper, I do not want to have to go digging to find out exactly how they came up with the results displayed in figure 1.
Also +1 to Benoit’s suggestion of rolling everything into one PDF, although I might suggest providing both versions on a publisher’s website. Sometimes I don’t really need the 60-page version with the supplemental information, the three-page paper will do. But I’d really like it to be available.
Supplemental figures are wonderful! How frustrating life was when papers were limited to the distilled bullet points (…Nature, Science…), never knowing how things were done because of a publishers word limits! The number of times I have found gems of technique, expression patterns, sequences, and informative control experiments in the supplemental materials leads me to disagree with reducing their use. Reviewers comments used to be answered by a private showing of additional data. Nowadays, these data can be viewed (and reviewed) by all. Hands off my sup. mats!
There is an important distinction between the casual reader and the specialist: reading a paper in my field, the more information the better! That can only encourage research as it disseminates data that are useful within the specific field. The casual reader might get a bit annoyed, but a good paper will communicate the story through the main body.
Obviously the issues of logistics (being able to download one PDF with everything, limiting the number of supplemental figures) are valid, but I would not support eliminating them.