Keywords/MESH terms versus subject-group(s)

[Melissa37]

Can people elaborate on what they think this refers to from our priority list?

This is what I think, using the eLife examples:
eLife uses article-categories for some specific "keywords" that are used in the navigation of article pages, with subj-group-type attributes. JATS provides no defined list of attributes, so we're all probably doing it differently!

<article-categories><subj-group subj-group-type="display-channel"><subject>Research article</subject></subj-group><subj-group subj-group-type="heading"><subject>Cell biology</subject></subj-group><subj-group subj-group-type="heading"><subject>Computer science</subject></subj-group></article-categories>

We also have keywords composed of the author generated keywords as well as a research organism from a defined eLife list:

<kwd-group kwd-group-type="author-keywords"> <title>Author keywords</title> <kwd> <italic>Salpingoeca rosetta</italic> </kwd> <kwd>Algoriphagus</kwd> <kwd>bacterial sulfonolipid</kwd> <kwd>multicellular development</kwd> </kwd-group> <kwd-group kwd-group-type="research-organism"> <title>Research organism</title> <kwd>Mouse</kwd> <kwd> <italic>C. elegans</italic> </kwd> <kwd>Other</kwd> </kwd-group>

We don't use defined MeSH terms, should we? This is medical, what covers all life sciences?

Please add your thoughts

[Melissa37]

See examples loaded to elements repo:

https://github.com/JATS4R/elements/blob/master/keywords.md

Please add your examples

[hubgit]

I've added a PeerJ example for subjects and keywords (and yes, quite different!).

[tommowlam]

RE: "We don't use defined MeSH terms, should we? This is medical, what covers all life sciences?"

If a controlled vocabulary like MeSH is being used as a pick-list of keyword or subject terms, then it could be useful for reuse for this to be indicated explicitly (but still leave the publisher the option of using their own list or no list)...
'kwd-group-type' is means to indicate what standard list has been used - but I dont think that is how it is used generally.

...so might be nice to have something more explicit for markingup where a list of terms had been used.

Oh, and we cant 'insist' on using MeSH or anything in life science, of course, as this will not apply to other disciplines (humanities, social sciences) or even physical sciences...

(I will add Ubiquity Press keyword example)

[Daniel-Mietchen]

Some thoughts on this:

• there should always be some indication of topic.
• I find it irritating that things like "Research Article" or "Insights" (i.e. publication types) get mixed up using the same tags as for real subjects like "Infectious diseases" or "Biochemistry"
• it would be helpful if topics that are hierarchically related (e.g. physics/ biophysics/ membrane biophysics) could be listed in a way that reflects that hierarchy (e.g. by referring to external ontologies)
• avoid "and" classes like "Microbiology and infectious disease" and use both of its individual components instead (i.e. "Microbiology" and "Infectious disease" in this example)
• avoid abbreviations in subjects/ keywords (e.g. "C. elegans" should rather be "Caenorhabditis elegans"); some things like "DNA repair" might be acceptable though

Pinging @rdrysdale

[rdrysdale]

Hi

The PLOS thesaurus (https://github.com/PLOS/plos-thesaurus)

• covers areas including but beyond Life sciences
• does not include "Article Type"-esque terms
• does extend beyond 'real subjects' to include things (like places e.g. Papua New Guinea or professions e.g. Librarians and organisms e.g. Acorn worms)
• avoids "and" classes where possible (it is not possible at top tier level) - these are not helpful in a taxonomy
• is pragmatic about abbreviations (though spells out Genus in those few that are valid terms :)) - but includes synonyms which are supported by our machine-aided indexing technology

it all depends what you are trying to do, with what
happy to contribute further thesaurus info/explanation if useful