Becoming a Literature Ninja

Post date: Dec 15, 2018

Being familiar with scientific literature is essential as a researcher – it fills the large gap between what you learned in class and what you can teach yourself through your experiments. I often feel like it’s the most important thing a scientist can do. But, generally speaking, it’s not like you can take a class on how to follow the literature, and so it can seem pretty daunting as a beginning graduate student or undergrad. Over the years, I’ve picked up a number of literature-related strategies, and I thought I would share some key things I’ve learned. This primer starts from the basics, and goes all the way to Advanced Literature Skills – hopefully there’s something in here for everyone!

1) Set goals. Despite its importance, reading the literature can often fall to the bottom of a list of priorities. For me, it helps enormously to set a regular goal of paper reading. I personally go for two papers a week – if it’s much more than that then I feel like I’m not reading the papers closely enough, but any less than that and I will be missing important papers. Find a balance that works for you.

2) Get organized. Inevitably, you’re going to accumulate a bunch of papers you want to read and need some way to keep track of them. Or you’re going to want to share a paper or refresh your memory of a note you took, and once you’ve read enough papers it can turn out to be surprisingly difficult to re-find the precise paper you’re looking for. I strongly recommend keeping track of papers electronically. I have really enjoyed using Mendeley (though there are a number of tools out there – I’ve heard good things about Zotero and Papers as well, but haven’t tried them, and I previously used CiteULike, which is web-only). You can keep PDFs in Mendeley and add comments and highlights. Mendeley works on tablets and desktop computers, and syncs between your devices and the web so you can access your papers and thoughts anywhere.

3) Read papers from your lab. Great, now you’ve got some goals and you’re organized – but what papers should you actually read? Start by looking up your lab on PubMed and finding the most recent and relevant papers. What’s great is that the authors of those papers sit right next to you! They’re going to be very happy to talk to you about their work and fill you in on anything you’re curious about.

4) Finding the important background science. A single paper can often turn into a treasure trove of other important papers to read. The introduction and discussion sections of papers are popular places to find citations of relevant and important past work – but don’t forget the methods section, especially for papers that use methods you want to learn more about.

5) Finding the important extending work. What about the opposite of finding the important work that preceded a paper – finding the important work that follows? A nice tool is Google Scholar: search for a paper you’ve read and click the “Cited by” link below. This gives you all the papers that cite the original one, and you can even search among those papers. Another option is Web of Knowledge.

6) Set up RSS feeds. New papers are coming out constantly, and RSS feeds are a great way to be notified about them. Basically, you collect RSS feeds (lists of articles) from different sources by adding their URLs to an RSS reader, which is effectively a website that acts as your personal inbox for papers. I use feedly as my RSS reader, but there are a lot of options out there. And where do you find the RSS feeds themselves? A lot of journals have dedicated feeds (here are the ones for Nature, Science, Cell, Elife, PLoS Biology, and bioRxiv Genetics, to name a few), but the most important tool is to generate custom RSS feeds in PubMed, as described below.

7) Create custom RSS feeds on PubMed. What if you want to be notified about new papers from a certain author, or on a certain subject? You can make custom RSS feeds on PubMed, and it’s amazing! Currently, the way you do this is by performing a PubMed search, and then under the search bar clink the link for “Create RSS.” Increase “Items Displayed” to 100 and click “Create RSS.” Done!

8) Impress people with your PubMed tag skills. Where this gets even more powerful is by using PubMed’s field tags, which also come in handy for everyday PubMed searches. You can search for papers from particular authors with [au], first or last authors with [first author] and [last author], or particular journals with [jour]. I also like to use tags to specify the types of articles that are returned (such as “review” [pt]). A handy term to add to a search is “hasabstract” [text], which filters out the annoying non-research articles (check this out: before and after). Another handy one is the affiliation tag, [ad], which can help specify authors with common names (before and after). Tags are extremely useful for creating highly customized RSS feeds (such as “CRISPR” and (“genetics” [jour] or “plos genet” [jour] or “nat genet” [jour])).

9) Get personalized notifications about relevant work. There are also useful services to get customized lists of new papers sent to your email, and I find that they work pretty well. You give them a library of papers (from Mendeley or another electronic organizer), and they’ll find papers they think are relevant to your interests. I like PubChase, and Mendeley also has a similar service. (Interestingly, I’ve found that they send me non-overlapping sets of papers that are both quite relevant.) If you have a Google scholar account, you can also see custom recommendations based on articles in your library (ie, your publication history), or get email updates of papers citing, or relevant to, an author you follow. A different simple option is to follow people on twitter who tweet out papers – this can be a good way to see other people’s thoughts on new papers, though I personally find it to be a time sink. There are also non-custom options, such as getting journals to email you their Tables of Contents.

10) Join your local Journal Club. Journal clubs are a great way to get the newest papers sent directly to your inbox, and to meet other experts in the field, and hear their opinions. If your community doesn't have a journal club that fits your interests – start a new one!