Research Remix

August 17, 2011

Citations in Supplementary Material can be indexed!

Filed under: Uncategorized — Heather Piwowar @ 1:52 pm

Many of us advocate for data citations in the references section of published papers, to facilitate tracking and reward through standard citation tracking tools like Web of Science and Scopus.  These tools aren’t able to track data citations yet, but they will be soon.

What other hurdles are in the way?  We need policies to get authors to include data citations in reference lists (feedback wanted on this draft policy).  But how feasible is this, really?  What about cases when authors use many, many datasets, should all of those really go in the article references section?  Can they? How do we handle physical limitations of reference lists for journals still printed on paper?  How can we get around the fact that citations in supplementary information are invisible?  I’ve gone on the record calling these issues elephants in the room of data citation.

I’m very happy to report that some of these issues have been overcome!  (Perhaps a long time ago.  I just learned the details this weekend.)  Kudos to Nature and Web of Science.  Nature includes only brief methods in its paper publications: Extended Methods are available online for many articles.  I’ve learned that if Extended Methods include references, these references are indexed in Web of Science.  (References in normal Supplementary Information do not receive this treatment and remain invisible).  Nature puts it this way:

References in Nature‘s Extended Methods sections, which are online-only but fully integrated into the full-text and PDF, are indexed in external databases such as PubMed. Supplementary Information for Nature, presented as a merged PDF online separate from the article PDF, does not usually contain references; see — Editor,Nature.

Note that references in non-Methods Supplementary Information do not seem to receive this same treatment at this time.  From their supp mat instructions:

Please note that we do not encourage deposition of references within SI as they will not be live links and will not contribute towards citation measures for the papers concerned. Authors who nevertheless wish to post reference lists should continue the numbering from the last reference listed in the print version, rather than repeating the numbering in the print version.

Here’s an example of the Extended Method integration at work.  I hope you are sitting down, it is pretty exciting :)

Behrends et al. Network organization of the human autophagy system.  Nature 466, 68–76 (01 July 2010) doi:10.1038/nature09204

The paper contains 36 references:

The Supplementary Information includes several Excel sheets with data, and also a PDF file with detailed methods.  The methods contain an additional two references:

Ok, here is where it gets cool.  The version in PubMed Central?  All 38 references.  The Detailed Method section refers to these references as 37 and 38:

And, finally, here it is in Web of Science.  All 38 glorious references:

Yay!  Those authors in the Detailed Methods get cred!

It demonstrates a way forward.  Not sure how scalable it is.  Let’s find out.  Journal editors?  If your Supp Mat references aren’t indexed by Web of Science, I’d try contacting them to ask how you can get it set up.  Does anyone know of other journals that index their supp mat?  Can someone with a Scopus subscription check to see if all the references are included there for this paper?

I’m off to go revise a bunch of slides about data citation challenges for future presentations.  Yay.

Thanks to Joel Hammond (Thomson Reuters) for pointing out this avenue in May, and to Veronique Kiermer (Nature) for recent detail on how it works.

1 Comment

  1. Hmmm… So they *can* count certain supplementary references but how many of them *are* counted?

    I tried another Nature article, published some time after the one you use as an example here, and I can find 2 uncounted, ISI-‘invisible’ references in it’s supp . materials.

    Ortega, F., Escaso, F. & Sanz, J. L. A bizarre, humped carcharodontosauria (theropoda) from the lower cretaceous of spain. Nature 467, 203-206 (2010). URL

    The invisible uncounted refs are (on page 2 of the SI):

    1.) R. B. J. Benson. A description of Megalosaurus bucklandii (Dinosauria: Theropoda) from the
    Bathonian of the United Kingdom and the relationships of Middle Jurassic theropods. Zool J
    Linn Soc-Lond 158, 882–935 (2010).

    2.) P. A. Goloboff, J. S. Farris, K. C Nixon. TNT, a free program for phylogenetic analysis.
    Cladistics 24 (5), 774 – 786 (2008).

    Now I’m not saying this is the norm; you have demonstrated an article that DOES, and I have provided an example that DOESNT… but I do wonder just how many Nature articles do have counted supp. refs. vs those that have ‘invisible refs’ and of course those that don’t have supp. refs at all (N/A).

    Perhaps a project for a student to further quantitate??? ;)

    It’s great that Nature have enabled this, but clearly submitting authors are still putting refs in non-Methods Supplementary Information, and therefore denying those refs citation counts. Can’t Nature just count all and any refs in *any* SI?

    Comment by Ross Mounce (@rmounce) — August 17, 2011 @ 3:42 pm

RSS feed for comments on this post.

Sorry, the comment form is closed at this time.

Blog at

%d bloggers like this: