Saturday, March 30, 2013

Journal Club


Every week I download scores of megabytes worth of the scientific literature to my computer desktop—papers that I have come across in a search, papers I think will be relevant to my research, papers in my field of study, papers that interest me because of their titles, papers that have cited mine. I don’t even come close to reading all of these papers, but I skim through many, and read a few.

I’m curious how people handle the massive amounts of information. I once asked advice from a senior colleague who told me “Read everything, and remember it all.” Yikes. He continued: “Failing that, read the important papers of the important people in our field.”

My group has a weekly “journal club” meeting in which everyone presents a five minute summary of a paper in the literature—preferably recent and project-relevant. The goals are to make sure we all keep up with literature of interest to our research, and to spend time analyzing how papers are written to notice what works and what doesn’t. To help with this I came up with a brief rubric for presenting papers:

1. Read a paper either from the very recent literature or an old classic
2. What is the take-home message?
3. Why did you choose this paper (i.e. Relevant to your research? Interesting title? High-impact journal?)
4.  Science: what was successful about this paper? What was unsuccessful?
5.  Figures: Was there a successful figure? What about it is successful? Any unsuccessful figures? Why/why not?
6.  Writing: what was successful about the way the paper was written? What was unsuccessful?

Please let me know about your experiences with journal club. Is it useful? What formats work? What doesn’t work? What are people’s techniques for structuring time in order to do a better job of keeping up with the literature?

5 comments:

  1. I think it's important to come away from reading a paper with more questions, either implicit within the text, or, better yet, original questions raised by the reader.

    ReplyDelete
  2. Thanks John--I will definitely add "what are my questions after reading this paper" to my rubric.

    ReplyDelete
  3. I suggest using a RSS (really simple syndication) reader to gather the TOC or ASAP articles from the journals of interest. This puts everything in one place and automatically gathers it as soon as they're made available on the journal's website. Google Reader is an excellent first choice, although they are phasing this feature out by summer 2013. Feedly seems to be where most are migrating due to seamless integration with your google account. Once you have the RSS of the journals then it's a matter of hitting the spacebar to quickly scan the title/TOC graphic of the aggregated articles. I have about 15 journals in my RSS which means about 100 ASAP articles per day. I fly through them on the first scan, saving anything that might remotely be of interest and then going back through that shortened list (maybe 10 articles by that point) to fully read the abstract. Of those 10 I might actually download the PDF for 2-3 per day. That whole process takes me less than 10 min per day. Then I generally set aside a block of time on Friday afternoon to go through those 10 or so articles I downloaded throughout the week. Done. The best part is that the RSS readers generally have very good search capabilities so that if you know you missed something by flying through too quickly on that initial sweep, it's easy to search for a particular author or topic later on. Where I've gotten into trouble with this scheme is when I don't keep on top of it and soon I have 500+ ASAP titles to filter through. It seems daunting but it goes quite quickly even with that volume. Keeping it a short daily task makes if far easier though.

    ReplyDelete
  4. Thanks Tony--I don't use RSS much--only for the occasional blog that I read. I have journals email me their TOCs and I mostly work out of my email inbox. Friday afternoon is good reading time for me too. Next question will be about managing references for papers & proposals. I use MSword for writing, and so far I don't have a really good way to deal with references.

    ReplyDelete
    Replies
    1. If your a Mac person, a program called Papers is great. iTunes-like interface with drag and drop capabilities. Drag a PDF off of your desktop into Papers and it automatically looks up the meta data and fills in all the journal information needed for bibliographies and referencing. It's good for searching (Web of Science, Google Scholar, PubMed, etc) organizing, reading, making notes and highlighting journal articles as well. Also, it ties in nicely with MS Word in that a simple shortcut key (ctrl, ctrl) pulls up a small Papers search box to quickly look up a ref in your library based on keywords, authors, etc and then easily insert those refs into a manuscript without having to toggle over to the full program. It'll then create a bibliography stylized to any journal type imaginable. I find it several orders of magnitude better than Endnote. Downside is it's not free but it's also not ridiculously priced for academia usage (~$40 I think). On the free and slightly more bare-bones side I have colleagues who like Calibre a lot. More freedom with reference styles and such but a bit less polished overall. Hope that helps.

      Delete