nbib file from PubMed : how to import

how do I import nbib files created by PubMed ?

do I have to do some other step ?
import directly from menu could not be done.

thanks
  • Zotero currently doesn't support nbib.
  • well why the hell not? pub med is easily the most used source for scientific literature deposition in the world!
  • First of all, please be respectful and civil.

    Then, Zotero does of course support pubmed, both via URL bar icon and per direct import of the pubmed xml such as http://www.ncbi.nlm.nih.gov/pubmed/8692918?dopt=Abstract&report=xml&format=text
    While sure, it'd be nice to also support the MEDLINE tagged format (i.e. nbib) I don't believe it offers any advantage over the xml so I don't see why it should be much of a priority.
  • The advantage is that Pubmed only supports batch exports of search results in the nbib format. This is the only way to quickly build libraries based on searches of indexed articles without having to manually add each one by hand (or vising the individual web-pages of each article). If I move into a new therapeutic area, I'll need to generate an entirely new library of references and the easiest way to do that is with a PubMed search. Furthermore, If I can batch import from pubmed into Zotero, I can set up scripting for bibliographic management of, say, weekly or monthly search updates. Zotero is a really great tool, and it's not Zotero's fault that PubMed only handles nbib exports for batch citations, but it is also really a no-brainer to include this file type as a supported import. For this reason alone, I will be forced to abandon Zotero and move to EndNote.
  • I assume you're talking about the "Send To:"-> "Citation manager" feature on the search result list, which creates an nbib file.

    Note that you can get the PubMed XML file of the same search results through "Display settings"->"XML"

    Finally, if it's a "no-brainer to include this file type as a supported import" and it seems like you know something about scripting and perhaps coding, you are welcome to create a translator for nbib file format and we'll gladly accept it.
  • also, if you want to script something like periodic search&download you'd want to use the API and not go through the webpage. Obviously, the API will export pubmed xml, too
    http://www.ncbi.nlm.nih.gov/books/NBK25501/
  • I agree that allowing nbib imports would serve a big subset of the user base (basically anyone in life sciences).
  • @darbyjack

    Since I've posted this comment, I've found that copy/pasting the XML from PubMed works just as well if not better than having to download nbib files. If you want to get citations from PubMed into Zotero, the best way is to choose the ones you want (with the checkboxes) and then go to the top of the list and choose display settings --> XML

    then copy all the text on the xml screen and go to Zotero and open the folder you want the items to go into. Choose file --> import from clipboard. Note that for very large lists (100 or more) this operation can take several minutes.
  • not quite clear what the problem is with pubmed's XML format? What do you need nbib for?
  • thanks for the responses. people email me nbib files all the time (typically hours before a grant is due!).

    I am able to use the XML approach for searches that I initiate.
  • I see. It's on my radar as my next larger project (it looks like OVID also relies on a version of nbib), but will probably take a couple of months.
  • fantastic! this responsiveness is a big part of what makes zotero such a useful tool.
  • @darbyjack

    In the meantime, if you have access to RefWorks, you can import an nbib file into your library, and then export it as something more useful (if you use Zotero for your workflow instead). Login to refworks and choose References --> import then use the dropdown menu to select Import Filter/Data Source = "NLM PubMed" and Database = "PubMed" and Encoding = "ANSI - Latin I"

    once the citations are in your library, you can export them into a refworks format and import that file into Zotero.

    I know its a bit clunky to work with two reference managers for this, and I don't know about other manager's (like EndNote) ability to read Nbib- but when I run across these files that's how I work with them.

    hope that helps!
  • edited February 5, 2014
    OK, I've put up a first version of a MEDLINE/nbib import translator. Turned out to be easier than I thought.

    Your version of Zotero will automatically update within 24hs, or you can update manually using the "Update Now" button in the "General" tab of the Zotero preferences. You need to restart Zotero/Firefox before the translator becomes active.
    The translator will only work with well specified nbib, as from pubmed directly.

    I can only do so much testing, so any reports of issues or errors would be great. Always post the .nbib file in question somewhere I can look at it, e.g. as a public gist on gist.github.com.

    Note that if you have the choiced, PubMedXML is always better.
  • fantastic, great news. will report any issues.
  • Probé usando RefWorks para importar formatos MLN PubMed (.nbib) y luego en formaro .RIS para Zotero y Mendeley y funciona!
  • Once you get the search results, change the search results FORMAT from SUMMARY to PMID. Then cut and paste the whole list into Zotero's ADD ITEMS (by identifier). It should populate the all details in your file.
  • Thank you @ richardagnes for that tip! Works like a charm and ends this whole argument. Still works for newly designed Pubmed.

    I must say I am really impressed with the functionality and completeness of this application and I thought Mendeley was good before.

    This is probably one of the most googled threads for new Zotero users. It may be helpful to post a "SOLVED" for OP's question at the top of the page so they don't have to follow the whole argument.
  • We don't typically mark things as solved, but just to be clear: in addition to adding items by PMID, the MEDLINE/nbib format has also been supported in Zotero for more than 5 years. We've recently updated to also work for nbib as produced by the ERIC educatoinal literature database
This discussion has been closed.