I was intending to write a post regarding an interesting paper on o-phenylene polymers. This paper describes experiments and computations on the hexamer, with particular attention paid to arene-arene interactions.1 These compounds fold into a helix, which has obvious application to many biological systems (DNA and the α-helix of peptides).
One of the things I am attempting to convey in this blog is the advantage of electronic communication in the sciences. In particular, I incorporate 3-dimensional structures of molecules in a way that allows the reader to interact with the molecule through a Java applet. (If you haven’t done this yet, any of the 3-D static images in this blog are actually linked to active structures – simply click on them and allow the Java applet to load.)
Now the paper by Hartley and co-workers does include supporting information with the coordinates of the different conformers of the o-phenylene hexamer, and I was all set to create images and incorporate the active molecules within a post. However, the pdf version of the supporting materials, while looking fine when viewed, actually has destroyed the data. I cannot copy-and-paste the coordinates into any program – the coordinates are completely corrupted! This is yet another example of how pdf is perhaps one of the worst choices for data deposition, as Peter Murray-Rust has often noted in his blog.
So until the supporting materials are fixed in some way, I will not, really can not, write up a post on it. Authors please remember to submit useful supporting materials!
References
(1) Mathew, S. M.; Engle, J. T.; Ziegler, C. J.; Hartley, C. S. "The Role of Arene–Arene Interactions in the Folding of ortho-Phenylenes," J. Am. Chem. Soc. 2013, 135, 6714-6722, DOI: 10.1021/ja4026006.
Henry Rzepa responded on 14 May 2013 at 7:48 am #
Like Steve, I often have to fight with PDF files, trying to extract coordinates from page-broken presentations without corruption. Like him, I sometimes loose the fight. In one extreme case, I retyped by hand (never again!).
So anything that comes along to make all our lives easier has to be welcomed. I have occasionally urged Steve to run a digital repository, where he could store coordinates (and other useful files) associated with this blog (on the grounds that extracting the coordinates from Jmol can be tricky. Steve, why don’t you update to eg Jmol 12, which as a signed applet has a file save dialog?). But running up such a repository is not simply 5-minutes work!
So I welcome Figshare, who have recently released a file uploader at http://figshare.com/tools. You have to also create a Figshare account, and thereafter uploading any data associated with eg a publication is really very easy. There is no need to squirt it into that PDF format any more.
What else might Figshare offer? Well, you can annotate any dataset with meta-data; for example an inChI key for data describing a molecule. You can add doi citations for relevant publications. All of these features are used by Steve on this blog; the idea is that they should also be present on any digital deposition.
Figshare also offer an API, and so it is quite easy to develop Web-based uploaders; we have been using several such for computational chemistry for a few years now (based on DSpace, Chempound and Figshare).
I would suggest that if there are any journal editors reading this post, you seriously consider urging your authors to adopt these mechanisms instead of simply submitting a PDF file of “supporting information”. I think it is going to take sticks rather than just carrots to get the community doing this.
Raphael responded on 17 May 2013 at 9:45 pm #
How do you usually plot the coordinates? Do you simply copy and paste from the PDF? What software do you use? I always have trouble copying from the PDF, and I always ended up converting the file from PDF to doc. But is does not always work. I think journals should provide a different plataform (other than PDF) for authors to copy the coordinates.