Tags/keywords with Imported Library

I'm new to Zotero, migrating our existing library from SharePoint. So far I have successfully drag/dropped batches of pdfs into Zotero and retrieved metadata. However, tags/keywords don't populate as they do when downloading from PubMed. Is there any way to get the tags/keywords without going back to PubMed? And if there's no shortcut, what would be the most streamlined approach to importing articles from SharePoint and matching them up with PubMed tags? Advice would be much appreciated.
  • Nothing super simple, but here's what I'd do:
    - Install the PM(C)ID Fetcher add-on for Zotero
    - Use that to ensure that all items have a PMID (note that the add-on requires a DOI, so items would need that, but most metadata from a PDF would)
    - Export all items as a CSV and clean-up the extra column so that PMIDs are listed in a single column
    - Use the add-by-identifyier (magic wand) icon to import from those PMIDs, about 100-200 at a time (more can get messy)
    - Then merge all duplicates (you can do this one by one or find a script to do so automatically on the forums here): That will get you the PDF from the original item and the tags from the PubMed item.
  • Thanks, Adam. After several tests of this process, only one out of 30 or so articles included a PMCID in the CSV download. Any guidance? And do you know if Zotero will be adding the tags in metadata retrieval in the near future?
  • That sounds wrong. Did you use the PMID fetcher add-on?
  • Sorry I'm not able to send screenshots. From the folder in my library... Select all... right click... choose "Fetch PMCID keys" from dropdown. (Nothing visibly happens) Right click... copy item to CSV.

    I just performed this exercise and 3 out of 26 items in the CSV came back with PMID numbers.
  • Ah, it looks like the fetcher will only work for items in PubmedCentral, not all of Pubmed. @emilianoeheyns any thoughts on working through that, e.g. by going through the full eutils API?
  • Doesn't look to be too hard, I'll see what I can do over the weekend.
  • Is it possible to add the keywords to the metadata retrieval and bypass the fetcher altogether?
  • No, the metadata comes from CrossRef which doesn't have MeSH terms. Also, not everyone wants MeSH terms.
  • Has anyone here worked with this API before?

    @vsimonnet can you open an issue on the github project? I need someone who actually uses pubmed to get this done.
  • I'd be happy to help get pmid functionality up and running.
  • Not as a techie, though, just a user and novice coder who needs pmid after leaving mendeley.
  • What I most need is people who can help me spec out the desired behavior, in issues on github, no coding required (although I'll take help if it is offered).
  • I was pulled off this project and just getting back to it. @emilianoeheyns and @adamsmith, did you find a solution? If not, I would be happy to work with you.
  • I have a vague recollection of doing something about that API a long time ago in the plugin, it might already work. If not, open an issue, and we'll see what we can get working.
  • Yes, the add-on now also fetches PMIDs of items not in PMC, so the above workflow will now work for basically all items.
  • I confirmed the current version of zotero and pmcid fetcher but my csv downloads don't include any pmid or pmcid numbers. @emilianoeheyns or @adamsmith, would you kindly list the steps in layman's terms so I can ensure it's not user error? Thanks so much.
  • If the PM(C)IDs show up in Zotero, the fetcher has done it's work. I'm not involved with CSV export.
  • @emilianoeheyns, PM(C)IDs do not show up in Zotero. I select all items in my library, right click, select "Fetch PMCID Keys." Nothing visibly happens, and no PMCID is added to the "info" tab of any item. The "modified date" also does not change. Please advise. Thanks.
  • For bugs with the fetcher you'll have to open an issue on https://github.com/retorquere/zotero-pmcid-fetcher/issues
  • And note that the items in question will need to have a DOI.
Sign In or Register to comment.