Dr. Isis is still trying to wrap her mind around Moveable Type, HTML, and all of this social networking that comes with being a totally hot, rockstar, science blogging diva. Still, every day I get a little better at it and I am starting, almost, to feel at ease with the whole process. I have recently learned to Twitter, which I thought was something one only does in the privacy of their own bedroom. Silly me! Dr. Isis is getting to be totally l337!
Figure 1: Dr. Isis is going to have to add "being l337" to the list of things she is goddess of.
So, knowing that I feel like I have entered a new realm and am trying to get my bearings, I could sympathize when I received the following letter (and, gosh, I really love reading your letters) asking for my sage and always spot-on advice:
Dear Dr. Isis,I come to you with a question. Do you have any suggestions for a young scientist (ok I'm not that young and I should have already developed this skill) to remain on top of her reading? How does one not let very important important science go unnoticed? Have you developed any tricks to keep abreast of the latest stuff that seems to come out at an ever-increasing rate???
Cheers to the wisdom of the Isis.
This is a fantastic question and, at first, I was unsure how to answer it. You see, my technique for staying abreast of the literature has changed as I have developed as a scientist. I imagine it will continue to change as I grow and mature and as technology changes. What I can give you are the tools that work for me now and the tools that worked in the past.
Figure 2: Dr. Isis's strategy for staying on top of the literature has changed as her career has changed. Also, these are the very first shoes I ever dreamed about.
When I began my career as a scrappy, and slightly perkier, younger scientist I came with no knowledge about the history of my field. Thus, understanding what had already been done in my field was very important. At the time, everyone lost their junk over PubMed and I spent hour after tedious hour trying to search for terms I thought were relevant, trying to see patterns in the authors names, and printing and filing paper after paper. At the time, this was the norm, but I always found myself in the situation where someone important would say something like, "Oh, come on. Didn't you read that paper by John Smith where he described the fundamental principle upon which our entire field, nay the entire world, was founded on?" To which I usually replied, "Fuck!!!" Or, I would be in a meeting and my mentor would say, "You know. That really important paper by that old white dude. Don't you have a copy of that?" Then I would have to try and remember if I had put it in the "old white dude" or "really important paper" folder and usually I would end up not finding it. I have papers that I printed out 6 years ago when I was working on a government-sponsored project, that have changed MRUs with me twice, in a pile on a bookshelf with a sticky that says "to be filed." I don't know why I keep them. I'm never going to file them. Perhaps I will send them to The Smithsonian.
After a few years of puttering around with that dinosaur PubMed, I transfered to an MRU with a subscription to Web of Knowledge. Web of Knowledge was like a little ray of sunshine in my life. It lets you follow an article's lineage in terms of the references is cites and the papers it's been cited by. Plus, you can enter a search term ("Why Dr. Isis is so Hot," for example) and map the authors that most frequently publish about that topic. Sometimes it's a dead end, but sometimes when you're getting into a new field it's a quick way to discover who the big players are. Then, I just quick export everything to EndNote and announce, "Voila!"
So, that's my trick for historical literature kung-fu. Now, as to the current stuff -- that can be a bit trickier. At this point, I am acquainted with a lot of the big players in my field and my knowledge of what they are up to comes from discussions, correspondence, or interactions at meetings and seminars. Still, I have four major sources of current reading material. First, some of my friends send me things they think I will think are cool. Most of the time they are, and that stuff ends up right in my EndNote under "Awesome Stuff My Buddies Send." Second, our large collaborative group has monthly journal club and each subgroup is usually responsible for a month's offerings. It's been a great opportunity to see what parts of the literature the people who work around me are interested in. It's also been an opportunity for me to be asked, "WTF crappy paper is this, Isis?"
Really, I know that journal club can seem like a hassle when you are bogged down with bench stuff, but it is a fantastic way to learn to discuss science Finally, I use two electronic methods to stay up on my science. First, all of the major publications I read (Journal of Applied Physiology, Chest, Circulation Research, Science, Horse and Hound) all have RSS feeds for their tables of content. I let them pile up and then on Sunday mornings I review them and read the articles that interest me. Some weeks there are many. Some weeks there are only one or two. The other electronic tool I is the RSS feed available from PubMed. You can create RSS feeds for your favorite search terms (like, "Dr. Isis is totally hot") and then articles which contain those terms will be sent to your favorite aggregator.
So, happy literature reading! If you have other methods of staying on top of things in your field, I'd love to hear about it in the comments.
Education & Careers
Technology
Comments
I use Google feed reader, notebook to tag stuff online I want to go back and review in detail later and homepage to stay up on the news, to-do lists, etc as soon as I log in. I used to use Zoho office, but I find iGoogle more user-friendly.
Posted by: literarydeadkittens | January 5, 2009 4:47 AM
I like to use a tool called Pubcrawler (http://pubcrawler.gen.tcd.ie/) that seems to work a lot like the RSS feed from PubMed that you described Isis. It also checks GenBank. You can set up multiple searches, set the interval for them to do the search, and have the whole thing delivered to your email inbox.
Posted by: Jenn, PhD | January 5, 2009 7:29 AM
In email conversation with a friend of mine who keeps up to date, they've said:
"Nature, NEJM, Science etc etc. most journals have weekly podcasts, which give you a summary of what is new and hot with respect to academic papers."
As they said, you can plug into them as you're doing round-the-house and round-the-lab, even play on the computer as you're working?
Since another blogging-friend of mine has pointed out in regards to getting a young man into science, podcasting is a new way of exposing one to ideas and online radio - it's something that the Skeptic Zone podcast is very, very aware of and we work on bringing in as much science news as possible.
Posted by: Podblack | January 5, 2009 7:35 AM
The only way to guarantee that you will find cool shit that you never expected to be relevant to your research is to browse tables of contents of as broad an array of journals as possible. By the time you are en established investigator in your field, keyword searching becomes completely unnecessary, as you always know what everyone is up to and you know it is going to be published in the journals you read already. Of course, if you are trying to access a field you are less familiar with, keyword searching is important, but only as a first step to find the key players and a couple of recent comprehensive reviews.
When Comrade PhysioProf was a wee pup, he used to make a daily trip down to the library after powering up a gel to browse the tables of contents of journals. Our library had a rack where they placed that day's arrivals of journals, and I would scan every single fucking one of them--even shit like "Horse" or "Hound"--while my gel ran.
Posted by: Comrade PhysioProf | January 5, 2009 7:53 AM
That was the story of my LIFE. Sometimes, it still is, but much more rarely. I've got a couple of tips though, for those of us who don't have Web of Knowledge.
For Pubmed, you can have a select list of keywords (as many as you like), and every day Pubmed will send you an update as to what's new with those keywords. It's not always completely up to date, but I often get stuff I didn't have that proves very relevent. I think you can include author names as well if you want to keep track of people.
Also, most journals allow you to get free electronic Tables of Contents sent to your email every time a new issue comes out. One of the profs in my field says we should be receiving at least ten journals worth. I'm not sure if I have that many, but I certainly have the main journals in my field (and the big ones like Science and Nature) represented. You just go through them when you get them to pick up on the latest and greatest. And it makes cruising for Journal Club articles (AND blog articles for that matter) a breeze.
Posted by: scicurious | January 5, 2009 9:20 AM
Because I want to track way too many journals, I have signed up for content alerts (most major journals have this option). I've got a hotmail account that receives all of the alerts, and then I can browse them at my leisure. This is especially useful for journals not carried by my library, and it saves me the tedium of having to check RSS feeds. Some publishers (e.g., Blackwell, Elsevier, etc.) even let you sign up for a whole buttload of content alerts at once.
Posted by: Andy | January 5, 2009 9:21 AM
HA HA HA..walk to the library. Sucker. Dr. Isis does not "walk to the library." She lets the library come to her.
Posted by: Isis the Scientist | January 5, 2009 9:32 AM
Long before literature search was computerized, ISI used to publish weekly a small booklet named "Current Contents." Later they put is on a 3.5" disks.
With all the good advice provided by Isis and the commenters, there is a portion of scientific literature that is found only in on dusty shelves in the library basement. You will be surprised how much science considered to be new was already performed decades ago. Journals such as Biochem. J. and J. Biol. Chem. recently electronized all their old volumes and put them on the net. I strongly recommend to do some browsing through those old volumes. You may discover that someone else has already done what you are just planning to do, or that what has been done can be either improved or used as a foundation for your future research.
Posted by: S. Rivlin | January 5, 2009 11:05 AM
These will probably be very useful to me, so thank you everyone. So far I had been blasting through PubMed and GoogleScholar with multiple tabs and keywords and searches at once. This led to me accumulating papers I want to read at a rate faster than I can read them. So lately I've just been sticking to my PLoS Pathogens, PLoS ONE, and PLoS Computational Biology email feeds.
I have a 2-ream stack of papers at home to read (the stuff I want to read) and a dozen or two papers for the lab that follow me around in my backpack with more filed away.
Here's a question from an aspirant scientist: How do you read papers effectively? I want to papers quickly, but I want to also retain them easily. Currently the papers I read wind up with doodles and flow charts all through the medians and it takes me about 1/2h per paper, longer if there's lots of math.
And a question directly at Dr. Isis: just how many EndNote folders do you keep?
Posted by: Toaster | January 5, 2009 12:17 PM
As a Grad student, I settled in on a workflow similar to CPP and scicurious that attempts to balance serendipitous finds with clearly relevant literature.
I subscribe to the table of contents (ToC) emails for about 8 of my key journals that I skim systematically, clicking through to and skimming 25-50% of the abstracts, and skimming maybe 1-5 articles in full text.
I subscribe to the ToC of another 20 journals of tangential interest/review journals (lots of Nature Reviews Whatever, Trends in Whatever, PNAS) that I skim lightly and click through to read maybe 5% of the abstracts.
Finally, I have about 10 PubMed RSS feeds and email alerts that search for arcticles by key collaborators or key competitors, or that mention a key gene or protein. Some of these are also set up to search PubMedCentral full-text, so if one of my favorite genes shows up in a microarray study, I hear about it 6 months after the article came out.
For me, the key to using email to manage ToC reading is to use Apple Mail to read the emailed ToC, and to hold the Command (Apple) key while clicking links. Then, the page opens up in Safari, but Safari stays in the background, so I can finish going through all of my ToC before switching to Safari to read things. Often, I do this on my laptop while on the bus to work. That way, I load safari with a bunch of links to follow, which would then load once I connected to the university network.
Maybe this is overkill....
Web of Knowledge sounds awesome. I'll have to figure out how to get access at my current institution.
Posted by: microfool | January 5, 2009 12:30 PM
Ah well, your tips are all good (and largely similar to my own strategies), but you didn't answer the question that still vexes me. How and when do you dedicate the time to actually read current papers? I have a very nasty habit of reading the TOCs and maybe the abstract and then setting the paper (electronically) aside and never getting back to it. Then when I get to the actual writing stage of a project, I discover that there's all this cool, relevant literature that I never actually read. doh!
Posted by: ScienceWoman | January 5, 2009 1:06 PM
Once you get good at it, you should be able to glean everything you need to know to incorporate new papers into your internal conceptual framework in just a minute or two. Of course, you can then return to the paper to extract more detail as needed later. The point of the initial 1-2 minute process is to enable yourself to know when you need to go back for more detail extraction in the future: such as when writing grants/manuscripts or planning/interpreting experiments.
Posted by: Comrade PhysioProf | January 5, 2009 2:02 PM
Woah woah - now you have a Twitter?
ah-ha. I like it when these things are easily found. Hooray!
Posted by: Chris | January 5, 2009 2:06 PM
I don't print out anything anymore. It takes up too much space, is heavy and is too hard for me to keep track of. If I can't get something in electronic form I don�t bother getting it at all. I subscribe to Science and Nature and read them as hard copy, but anything important for my work I store as a pdf in my filing system.
I don't use any fancy file system; just file folders with pdfs of the paper. I give the pdf a name that is reminiscent of the title (usually the first words or the whole title if it fits). I put these into file folders with an idiosyncratic hierarchical structure and an idiosyncratic name. Early on, I would put papers in multiple folders so they would be easier to find, but I have stopped doing that because it takes up too much space.
I am not at an institution that has any subscriptions or a library, but the school I went to (MIT) allows alums to use the library. Periodically I make the trip and download everything I can while I am there. My philosophy is that if it is worth looking at the abstract, it is worth having an electronic copy of. I bring them back, store the file on my computer and gradually integrate them into my filing system.
At home I use PubMed, at MIT I use Google Scholar and PubMed. PubMed is useful for getting older papers which are very useful. I always look at forward and backward citations and get them if they look interesting and I am able to.
Because I can't get everything (there are lots of journals that MIT doesn't subscribe too, lots of medical ones), I try to get "enough", that is enough so that I can fill in the gaps between what I have. You always have to have that skill anyway, because if you haven't read a paper, you might still need to figure out what is in it. Sometimes there are no papers when you go looking for those that "should" be there. They are not there because they haven't been written yet and so you can be the first. That skill is also useful when you need to decide which papers are crap and which ones are absolute gems. Sometimes the authors don't know which is which.
The question of when do I have time to read everything, I don't. I read very fast, even just a minute looking at a paper will give me a gist of it, which is "enough" for background and to put ideas into the back of my mind. If I want more details, I always go back to the paper and look at it more carefully. I have enough trouble remembering my own name, remembering the names of old white guys is not something I can do. I can remember their work, and often have a copy of the paper, but arbitrary details like names are much more difficult for me. The answer to the question of "when do you know something", is, "when you can use it".
Right now my top most folder (titled "papers") has 67,744 files, 3,875 folders and is 20.2 GB. Some of those are duplicates, and some are web pages which get stored as multiple files. This is small enough that I can load it on my laptop and take it with me where ever I go. That also includes my experimental data and all works in progress, my blogs and blog comments.
I have thought about going to a different system, but I don't consider myself very computer savvy (the conventions are too arbitrary). What I have works, is scalable, is portable, and is free. Information is only useful if you are able to access it mentally. Sometimes I do forget what I have and only rediscover it when looking for something else. My main constraint is getting the paper, not the time to read it or understand it. If I can't get it, I can't read it and so can't cite it. Open access is great.
I very much agree with CPP that reading papers across disciplines is extremely important and useful and also that once you get good at it it only takes a few minutes to incorporate it into your conceptualization. I have found that if I have figured something out which changes my conceptualization, the implications of that change only become incorporated into specific areas while I am actively thinking in those areas.
Google has a searching capability that you can download and use to index your computer. I use that when I can't find something that I am pretty sure that I have. That is quite useful when I have a backlog of unfiled pdfs (which is always).
Posted by: daedalus2u | January 5, 2009 3:36 PM
As a librarian, may I suggest one cool way of finding more unusual literature that may be relevant to your field? If you have access to Web of Knowledge or Scopus, do a weekly (or monthly) citation search/alert on yourself - if you have a decent body of work you may find yourself cited in journals and on research you might never see otherwise.
I miss serendipity...and print journals...and are librarians ever hot?
Posted by: Patricia | January 5, 2009 4:07 PM
Isis!! Those were the first shoes I ever dreamed about too!!!! SHUT UP!
@Patricia: Librarians are hot, if not in real life than in the fantasy lives of men everywhere.
Posted by: Arikia | January 5, 2009 4:58 PM
Patricia, Arikia is right-- librarians are totally hott. Especially when they wear those black framed glasses women are wearing these days.
Posted by: Riesz Fischer | January 5, 2009 5:37 PM
I once knew a very hot librarian, and her hotness was no fantasy.
Posted by: daedalus2u | January 5, 2009 6:45 PM
Alright, I am going to speak authoritatively on this shit. Librarians are sexy as hell. Seriously, I just bought a skirt and blouse to give a seminar in and thought to myself as I tried them on, "If I forgo my contacts and wear my glasses with this, it's totally naughty librarian."
Posted by: Isis the Scientist | January 5, 2009 6:47 PM
...
.
...tom...
'who is loving the way this comment thread has morphed to hott librarians. 'Fantasy lives of men everywhere' indeed..!!'
.
Posted by: ...tom... | January 5, 2009 6:57 PM
Hil-fucking-arious.
Posted by: gnuma | January 5, 2009 8:02 PM
Like Sol, I still love rifling through the moldy stacks of pre-1966 journals not online. My mentor was fond of saying that two hours in the library saves two months in the lab.
Dr Isis may be tickled to learn that I used the original 1948 Ahlquist paper from Am J Physiol to explain to the assembled masses of eager young minds how α and β receptors were distinguished by differential rank order of potency of a phenylethylamine panel in various organ preps. Now *that* is hot...plus Ahlquist was such a hipster he was in Athens long before the B-52s and R.E.M.
Oh yeah: library science and librarians? Love 'em, can't live without 'em.
Posted by: Abel Pharmboy | January 5, 2009 8:11 PM
OK, I've just got to plug some ScienceOnline09 hotness!
Drop by the wiki and let Christina and I know what you want us to talk about.
Sadly, I'm an oldish white dude, but Christina isn't!
Posted by: John Dupuis | January 5, 2009 8:26 PM
Toaster: yeah, I feel you on the paper reading. I know when I was a first year grad student, it would take me ALL DAY to read a paper and feel like I gleaned something, because I didn't have the background necessary to pick it up. Now that I do, I can read a paper and take notes in about 20 min.
Here's what I do:
1) Find somewhere very, very quiet. Set yourself there for a previously decided period of time. No excuses. Since it takes me about 20 min to read a paper, I go for 40 min. time bins, say, when something in the lab is running. I do NOT recommend being within reach of the internet.
2) I keep sets of notes. For me, this works best when I physically write things down. I have notebooks that are set aside for reading notes, with tabs on each page noting the author, the year, and the general gist of the title.
3) Take notes, these can be sloppy. Include diagrams. For me, on a topic I know pretty well, each paper will come out to about half a page of notes.
4) Keep either another notebook or a VERY well organized folder. Refer to this folder as your "thesis bible". I do a notebook (which is scanned periodically so I have an electronic copy and don't lose my shit if it gets lost). Each page is a tab with a paticular aspect of my research. Examples include headings like "dopamine and GABA", "dopamine and Glutamate", and "dopamine and your mom". In each of those sections, I take the gleanings of my notes on the papers. I try to distill each paper down to one sentence. Then, in the side margin, I write the chapter and verse (author and year) for who found the thing I'm looking for. I also include diagrams when I feel they really explain the system.
This sounds complicated, but it makes putting things is endnote or reference manager ridiculously easy, as you know exactly what you're looking for, and you have all things on a particular topic (for, say, a particular endnote folder) in one little column. Not only that, when you go to write a paper, and you say "hmmm, dopamine and your mom, what do I know about dopamine and your mom?", just flip to your section, and there it is, in neat little columns with references, everything you know about it.
So that's what I do, and now we ALL know how obsessive Sci is. But she's organized!
Posted by: scicurious | January 6, 2009 10:28 AM
I'm really going to have to stop reading this blog at work. I get too..."bothered" is a good term, I think. Nice, neutral term. Not exciting at all.
Posted by: Ranson | January 6, 2009 3:03 PM
I tried listening to the NEJM podcast a couple of times.
But damn!
They speak so slowly and are so lacking in animation that by the end of the sentence I wonder whether someone has given them after school detention and have mistakenly given them the NEJM press release instead of "I will not speak out of turn in class" to read out.
Posted by: nm | January 6, 2009 5:17 PM