Filtering tools for the scientific literature?
April 29, 2013 4:45 PM   Subscribe

I'm a research scientist (in biology, focused on microscopy), and I'm looking for new tools that will point me towards relevant research papers. I've been playing around with Readcube and its recommend feature, which looks good, as does Google Scholar's 'updates', and I have various journals send me their table of contents, but I feel like there must be better tools out there to point me towards new research that I should know about. Particularly with the ever increasing number of papers being published in journals like PLOS One, I'd like to find a better way to pull out the few that are relevant to me from the torrent of new data being published. It seems like there must be novel tools to do this out there, but what are they?

I run a core research laboratory for the university and so I'd like to be kept up to date on new techniques and new methods. Also welcome are tools for organizing papers. I use Zotero right now and have been playing with Readcube, but other suggestions are welcome.
posted by pombe to Science & Nature (19 answers total) 27 users marked this as a favorite
Not sure if it's quite what you're looking for, but I saw a presentation recently about ScienceScape that makes me think it might be useful.
posted by arco at 5:19 PM on April 29, 2013

Don't know which specific article databases your university's libraries use, but most academic databases offer a feature that will automatically RSS or email you a copy of new articles that match a search criterion. A librarian should be able to advise you as to the best way to get what you want in terms of the search keywords to use.
posted by Rykey at 5:19 PM on April 29, 2013 [2 favorites]

Seconding Rykey. There's likely a librarian at your university who specializes in your field. Find that person, and make an appointment with them to ask exactly this question. They can tell you about database features, and may even offer some sort of alerts service via the library. They may also have suggestions and tools for organizing your research.
posted by donnagirl at 5:35 PM on April 29, 2013 [2 favorites]

If you'd like to pursue that route, I'm making some guesses, but I think this is probably your librarian.
posted by donnagirl at 5:40 PM on April 29, 2013 [1 favorite]

I agree with the above. Look on your university library's website for the research librarian for biology, or the reference librarian for the sciences. Or just ask the reference librarian who answers the reference phone or chat who you should contact.
posted by mareli at 5:41 PM on April 29, 2013 [1 favorite]

A bunch of people in my field have started using ResearchGate as a collaboration/networking tool. It's a decent way to keep track of what people are publishing/reading/working on.
posted by no regrets, coyote at 6:07 PM on April 29, 2013 [1 favorite]

Are the journals you read indexed in PubMed? You can set up alerts by subject, author, etc. in PubMed (your librarian can show you how if you can't figure it out). You can get a lot more granular with something like PubMed than you can with Google Scholar.

Actually a lot of databases have that feature; see what your institution offers.
posted by mskyle at 6:08 PM on April 29, 2013 [1 favorite]

Database searches are not what I'm looking for. I have set a bunch up in google scholar for specific things I'm interested in, and probably search pubmed every day, but it's not solving this problem.

For instance, one thing I'd really like to know is every time an improved fluorescent protein is published, but I don't want the much larger number of papers just using fluorescent proteins. I'm also interested in things I should know about but hadn't even thought to set up alerts for - the kind of thing a colleague mentions in passing - did you see the new cool paper / cool technique / etc.

This question was in part inspired by this recent Nature news piece - it seems like there has to be good tools for leveraging the wisdom of crowds here (people who read the papers you think are important also read these papers) or for leveraging machine learning (these papers have similar content to papers you think are important).
posted by pombe at 6:14 PM on April 29, 2013

Your update doesn't change my mind, I still think a good academic librarian could help you refine your search to give you more of what you're looking for, especially in the case of your "for instance" example. If database searches aren't working for you, a librarian is *exactly* what you need.
posted by donnagirl at 6:29 PM on April 29, 2013 [2 favorites]

Mendeley does recommend similar articles (I think both by content and by what other people read) in the web interface, but I seldom actually use the web interface, so I have no idea if it does do a decent job.

Everyone I know solves this problem by having the arXiv email them every morning and then training themselves to actually read the email. But I can't find an analogue of the arXiv in your subject, so that doesn't do you much good.
posted by hoyland at 6:33 PM on April 29, 2013 [2 favorites]

I'm with donnagirl. Talk to an academic librarian in your field.
posted by Rykey at 6:42 PM on April 29, 2013

So, not knowing your field - how often is a paper with a new improved fluorescent protein published? Is this like a weekly thing or a monthly thing or a "once every couple of years" thing? Are there words that often come up in the title or abstract of such a paper (like "novel" or something)? These are the questions I would be asking if I were your librarian.

(That Nature news piece is full of great stuff, but unfortunately that kind of thing has taken off much more quickly in physics and math than it has in biology - hence hoyland's comment about arXiv.)
posted by mskyle at 6:54 PM on April 29, 2013

As others have said, talk to your librarian about your specific situation. Also I'd strongly recommend sticking with Zotero.
posted by singingfish at 6:59 PM on April 29, 2013

I'm going to second hoyland and no regrets, coyote in the half-recommendation of Mendeley and ResearchGate. If you can get people in your field using it, or find people with similar interests and possibly more connections (people who act as referees for these sorts of papers, people who go to lots of conferences, etc), you'll have a good sense of what's having an impact in the literature sooner.
posted by 0x006DB0 at 7:14 PM on April 29, 2013

Google Scholar Alerts for anything you can express as keywords; Scopus also has an alerts service. ResearchGate seems promising if lots of people in your field are on it and you can follow the most relevant ones; I've only joined recently so haven't really evaluated it thoroughly yet.

One technique I've had recommended (but haven't actually tried) is to use Twitter. A surprising number of researchers twitter about science stuff; if you can find any whose interests match yours, follow them and you have free notifications of stuff you might like.

I don't think there's a magic bullet here. To address your fluorescent protein example: I don't know of any semantic search that could filter out only papers about new fluorescent proteins while ignoring ones about applications of known fluorescent proteins. If I wanted to do this, I'd just set up a Google Scholar alert for "fluorescent proteint" (and/or other relevant keywords) and filter the results by eyeball. Perhaps you'd get a huge number of false positives, but it doesn't take long to read the title of a paper.
posted by pont at 10:06 PM on April 29, 2013 [1 favorite]

This doesn't answer your question, but Twitter. I get a lot of this on Twitter. It's becoming a huge trend in academia in general, but depends on the size of your field and how many scholars from your field are on Twitter.

In a study I worked on in 2011, we found that 1 in 40 scholars was on Twitter. This number is undoubtedly larger now. 60% of what they posted we classified as "non-scholarly," but 40% of the rest of the content was directly related to their academic pursuits, including research. A small percentage of these are citations to research. In that second linked study (which was actually conducted first), we also talked to a bunch of scholars who were on Twitter about why and how they used the service.

Worth looking into, to see if it might be worthwhile. I know that I find it personally valuable, and many others do as well. A big part of the value is not as much in the citations to literature, although you can follow journals who will push the current table of contents to you (you can usually do that via email as well; I do this and scan the titles of articles I want to read) - but Twitter does foster a sense of community, gives you a bigger web of contacts, and an awareness of the field that I think is really nice, at least for me.
posted by k8lin at 11:46 PM on April 29, 2013

My thought is that you need to devote some time to manual searching and setting up your own alerts. You can't rely on a librarian, although they are certainly very helpful. But there's an aspect of this that requires that the range of search terms and potential alerts evolves as you become more familiar with your field (not to say that you aren't already!). My job requires me to search in areas that I'm only partially familiar with and this approach works well for me.
posted by gilast at 5:20 PM on April 30, 2013

Journal Lab is a project out of UCSF that may be interesting. You create a "dashboard" of pubmed search terms, and the system notifies you as new papers are posted to those feeds. You can then download them, put them in folders, etc. It's useful if you want to track lots of terms without cluttering your inbox.

(Full disclosure: I'm the developer on this project. It's getting pretty heavy use at UCSF.)
posted by davidgljay at 5:34 PM on April 30, 2013 [2 favorites]

I just encountered Research Lei:

This is a free, open source, BSD-licensed tool that allows you to build a library of papers that are relevant to your research and to easily explore the related work. Additional tools include search, a listing of often-referenced papers you may have missed, and a graphical visualization of the main researchers in your area and their co-author networks. Structured data is scraped using Microsoft Academic Search API.

Still in beta. Intriguing.
posted by unknowncommand at 5:28 PM on May 5, 2013 [1 favorite]

« Older Have rods, tackle & permit, will travel near...   |   How do I tell my parents I am taking a break from... Newer »
This thread is closed to new comments.