Citation profiles for our NAR and Concise Guide papers

Introduction  (this is an update from the original Feb  2017 post)
Like other database resources, we collate measures of impact for various uses. These include our own internal six monthly  IUPHAR/GtoPdb review meetings, documenting outputs of past grants and applications for new ones. While there is a widening choice of these metrics  (including website accesses, data downloads, hits on Open Access papers, papers linked to grant numbers and Altmetrics scores) citation counts remain an important component, while acknowledging contravertability. As of Jan 2018  we are preparing our application to become an ELIXIR Europe Core resource (following on from our successful incorporation into the ELIXIR UK node). In addition, these metrics should be useful for other funding explorations this year.  What we would like to explain here, underpinned by bibleometric analysis, is why some of our citation counts appear anomalously high (notwithstanding the fact that different sources generate different numbers anyway).

As a preamble,  it is clear that PubMed (including PubMed Central, PMC) and European PubMed Central  (EPMC,  see PMID 29161421) now have divergent but complementary feature sets for abstracts and PMC full texts (e.g. Entrez and MeSH x-refs on the NCBI side vs EBI x-refs and text-mined biological concepts on the other, also EMPC with 10 mill more abstracts).  This means that  anyone wanting to explore bibliometrics is now faced with a two-stop-shop (yet another de facto transatlantic split). This results in a certain amount of hopping  depending on what filters are needed. One of the quirks is that EPMC citations are significantly lower than for PubMed, Clariviate WOS or Google Scholar. Taking PMID 24234439 as an example, the figures are 440, 678 and 736, in that order.  While EPMC’s state their lower metrics are based on open citation data rather subscription-based services there may be other factors involved.

Nucleic Acids Research Database papers
Including our previous incarnation as IUPHAR-DB, the Edinburgh team and their collaborators are proud to have just had our sixth paper in Nucleic Acids Research Annual Database issues as listed below. Two sets of citation counts from EPMC are included( as of January 2018 so clicking should notch up additional counts since then . The reasons for two sets are explained below.

Up until 2013 we consider our citation counts for those  three earlier papers (depending on comparative benchmarks of course) as respectable. However, the ~10-fold citation jump for our 2014 and 2016 efforts are clearly anomalous. The causality can be discerned from the appropriate intersects and  diffs to generate the second column of citations. This shows the majority (in the left column)  are coming from the British Journal of Pharmacology (BJP) and British Journal of Clinical Pharmacology (BJCP).  This reason for this is because our NAR papers for 2014 and 2016  are selected as one of the reference citations derived from the Tables of Links (ToLs). These valuable connections between ligand and target entities specified by authors and GtoPdb records, were initiated in November 2014 (see “BJP is linking its articles to the IUPHAR/BPS Guide to PHARMACOLOGY” PMID 25965085) .

By 2017 three changes had been introduced.  The first was the ToLs were moved to being “in-line”  outlinks at their first mention in the text.  The second was that they began to be included in BJCP as well.  The third was that, while for the first two years BJP Wiley Editors added the ToLs references  to the author citation lists as reference citations post acceptance, this task is now covered by authors. This is described in the BJP and  BJCP instructions for authors,  a section from which is shown below.

The selected references have thus sequentially included the 2014 and 2016 NAR papers in the table above,  along with whichever reviews from the Concise Guide series best match the theme of the specific article.  Since the ligand and target links are included in nearly all  BJP papers (but fewer in BJCP) these have rapidly accumulated citations. These reference publications will be updated to the 2017/2018 NAR and Concise Guides.

The key points from the table above are b)  the mandated reference to non-mandated reference ratio is ~10:1 and b) the citations for the latter are still very respectable. An interesting (but somewhat frustrating) aspect is that even the older NARs continue to slowly acquire non-BJP/BJCP citations from chronic “citation lag” by certain authors.  For example the 89 cites for our first NAR in 2009 (PMID 18948278) includes three that we appreciate – but – the authors ought to have picked up the 2016 database description (i.e. an eight year lag).  This effect is clear from the graphic that is now a usefuly in a right-hand facet of all EPMC abstracts.

We can a similar smearing of citation counts over the whole set of NARs  (e.g. high rates of  2016 papers citing the 2014 NAR).  So, for the record, anyone reading this who is considering citing us (elsewhere from BJP/BJCP)  is herby encouraged to use the 2018 reference (n.b. the BJP/BJCP reference citations will not have such a pronounced lag since they are systematically updated)

Concise Guide citations
To get an overview of BJP internal reference citations (as opposed to the external NAR papers shown above) we can view their  24739 papers ranked by citation count. First and third places are two successive editorials on “Animal research: reporting in vivo experiments: the ARRIVE guidelines” as PMID 20649561 , with 1130 cites and PMID 20649560 with with  905.  The Guide to Receptors and Channels (GRAC) series is topped by GRAC 5 (PMID 22040146) at 547 citations.  The Jan 2018 citation counts for the top seven of the 30 papers from the Concise Guide series (i.e. three sets of  over six years ) are  shown below.

As is implicit from above, since the introduction of ToLs, the “Concise Guides” are now also reference citations (n.b. depending on the time elapsed from Jan 2018 the live citation numbers above will have increased in the links).  Consequently, the total: non-BJP/BJCP citation ratios are similar to those for the NAR articles, as in the case of  PMID 24517644 where we see 444 : 50.  Unsurprisingly, GPCRs and Enzymes are the most popular while Catalyic Receptors and Overviews have the lowest citation figures.

N.b. the latter part of this original Feb 2017 post included discussion of  PubMed Commons and Altmetrics but this has now been split out into a new post. Additional aspects on the citation theme above are expounded in another  blog post.

Chris Southan, Jan 2018