Bioinformaticians (and anyone else who programs) love effective automation of mundane tasks. So it may amuse you to learn that I used to update PMRetract, my PubMed retraction notice monitoring application, by manually running the following steps in order:
- Run query at PubMed website with term “Retraction of Publication[Publication Type]”
- Send results to XML file
- Run script to update database with retraction and total publication counts for years 1977 – present
- Run script to update database with retraction notices
- Run script to update database with retraction timeline
- Commit changes to git
- Push changes to Github
- Dump local database to file
- Restore remote database from file
- Restart Heroku application
I’ve been meaning to wrap all of that up in a Rakefile for some time. Finally, I have. Along the way, I learned something about using efetch from BioRuby and re-read one of my all-time favourite tutorials, on how to write rake tasks. So now, when I receive an update via RSS, updating should be as simple as:
rake pmretract
In other news: it’s been quiet here, hasn’t it? I recently returned from 4 weeks overseas, packed up my office and moved to a new building. Hope to get back to semi-regular posts before too long.
