Mendeley is a pretty nice (and free!) reference manager that syncs to an online account (currently with 2GB of storage for free accounts). While it imports PDFs as citable references with remarkable accuracy, and is wonderful for a cite-while-you-write workflow, it doesn't always pick up PMIDs and PMCIDs. Certain citation styles, such as those used by the NIH for grant proposals, require you to add these IDs to your bibliography.
This little script attempts to fix that. It scans your Mendeley database for journal articles and does the following:
- If the PMID is available, it is used to retrieve the PMCID and DOI.
- If there is no PMID, but the DOI is available, it is used to retrieve the PMCID and PMID.
Whatever can be retrieved that is not present in your database is updated accordingly.
Bonus Feature: Mendeley has an annoying bug where if an author's first name(s) are fetched and they include the initial "A", it gets saved as a lower case "a", producing citations like "Smith J a" instead of the expected "Smith JA". This script fixes those names as well.
The PMC ID Converter and EUtils APIs are used to try to match your IDs, in that respective order.
Read this carefully. Although this tool is fairly paranoid and backs up your Mendeley database before touching it—every time it runs, as a separate file into backups/, even on dry runs when no data will be written—it does operate outside of Mendeley's public APIs for accessing your data. It could very well blow away your database or overwrite it in ways that you do not like. At some point I may rewrite this to use Mendeley's API so that it works on your online account's data, but that would require me to set up OAuth2 and a whole lot of Not Fun things, so no go for now.
If losing data concerns you, and it should, backup your Mendeley library before using this. You can do this by running Mendeley and looking in the Help menu for Create Backup...
The script operates locally on your Mendeley database, which ends in .sqlite and is named after your Mendeley username, and will be found in one of the following folders:
- Mac OS X: ~/Library/Application Support/Mendeley Desktop/
- Linux: ~/.local/share/data/Mendeley Ltd./Mendeley Desktop/
When you make a backup as previously described, Mendeley hands you a ZIP archive of this folder, which will also include all the files for your imported articles. This script simply copies the aforementioned .sqlite file into a backups folder before it operates on it.
Linux and Mac computers are supported. You'll need Ruby ≥1.9.2, RubyGems, and some basic gems, which will be installed with Bundler.
On most Macs, you'll only need to run sudo gem install bundler because you already have Ruby 2.0.0 and RubyGems, as of Mavericks (10.9).
On Linux, I defer either to your package manager or your raging desire to build from source.
In both cases, rbenv might help if your system Ruby is out of date.
First, clone this repository to a directory and cd into it. Then:
$ bundle install
to install the required gems. Bundler may prompt you for your password if it needs to write to a system directory. Of the gems needed, Nokogiri is probably the most finicky to install, so you may need to refer to its installation instructions if it cannot compile a certain library—but usually it can do it without intervention. Then:
$ rake
will perform a dry-run of the matching procedure. If everything goes well, you will see the script start running through journal articles in your Mendeley database, reporting if it can cross-match them using the aforementioned APIs. If it looks like it is doing a good job, you can stop the process with Ctrl-C, then:
$ rake update_ids
will start saving these IDs to your Mendeley database. Technical Note: because Mendeley has no dedicated PMCID field, the PMCID is mapped to its CSL variable by jamming it into the Notes field as specified here. From there, it can be parsed out by citeproc-js to be included in your bibliography.
These tasks will not run unless Mendeley is shut down, and every time it runs, you will first see a new .sqlite file backed up to the backups/ folder in this repo, just in case something goes horribly wrong. In the rare event that this occurs, remove the numerical prefix and copy it back to the appropriate folder listed in the Fair Warning section of this README.
The next time you start Mendeley, you will see your new DOIs, PMIDs, and PMCIDs, and it will automatically sync them to your online account, so your other computers will receive these IDs too. Note: I've had some trouble syncing the PMCIDs, since it doesn't seem that the Notes field is reliably synced. This seems to be a common Mendeley issue. You may need to run this script again on other computers if you need the PMCIDs there.
If you want to use a citation that prints these IDs after your references in your bibliography, check out this CSL. You can install it for Mendeley Desktop by going to View > Citation Style > Journal Abbreviations ... changing to the Get More Styles tab, and pasting the URL at the bottom.
The MIT License (MIT)
Copyright (c) 2014 Theodore Pak
Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.