Biocurious

/ a biophysics blog

SciBarCamp report: Digital lab notebooks

Posted 16 March 2008 by PhilipJ under &

The first SciBarCamp is going on this weekend at Hart House at the University of Toronto. The basic idea of SciBarCamp is that of “a gathering of scientists, artists, and technologists for a weekend of talks and discussions.” The kinds of things that have been discussed this weekend are the forefront of science (quantum gravity, synthetic biology, open access, scientific software), and the interactions with science and technology and art. There’s an extremely interesting mix of people here today, from quantum information theorists from the Perimeter Institute to social scientists from OCAD, grad students, science writers, musicians, etc. Quite the eclectic group.

The “un“conference opened up with what was perhaps the most interesting discussion for me. It was by Corie Lok from Nature Networks and John Dupuis from York University on “Science 2.0”, or basically how technology is changing first the way we do science, and then how we publish and track science. Things like Connotea and del.icio.us were discussed as ways to keep track of relevant journal articles, but the discussion was dominated by discussion about keeping track of your experiments, and how this is changing in laboratory environments.

Many labs are moving to an all-digital lab notebook. Gone are pens and paper as the primary means of keeping track of your experiments, and they are instead being replaced by digital equivalents, such as wikis or some other, commercial software (which I unfortunately didn’t catch any names, but a google search brings up all kinds of examples). The advantages of such a “notebook” are clear: entirely searchable, easily referenced with hyperlinks, the ability to keep digital images and snippets of code, etc. There is also an advantage which hadn’t been discussed, but is something I’ve long thought about: legibility! If you’ve ever looked in another scientists notebook, you will quickly find that their writing may be atrocious (guilty as charged), and it quickly becomes an exercise in frustrating trying to interpret what was written.

Perhaps not surprisingly, this is being embraced in a big way by the big pharma companies, where it is crucial to have detailed records, and to keep them for a long time (something like the lifetime of a drug being sold + ten years). But there’s an advantage to graduate students in academic laboratories that hadn’t been addressed in the discussion, that came to me as I was talking about this to another grad student friend. It’s that of the copies. I graduated from SFU with my Master’s less than a year ago, but I could tell you basically no specific details on some of the more mundane, day to day experiments that I carried out. I’m sure the same thing happens to those who graduate with a PhD, and leave their lab notebooks behind. I’ve decided I want to keep this kind of information, and I’m still early enough in my PhD that it won’t be too hard to change.

To the other scientists reading this blog: have you changed to a digital lab notebook? Have you tried the switch and failed, or have things gone smoothly? What software did you decide on? I’m leaning towards a wiki, and I’ve decided it will be necessary to keep a paper notebook as well, but to make it degenerate information which will, at the end of every day, always end up in the digital notebook as well.

  

  1. Mad Omicist    16 March 08    #

    I’ve gone with a wiki at one of the labs I’ve worked in. It went quite well, but the main problem I encountered was that it was hard to consult my notebook or jot quick notes in it during experiments. It wasn’t a total deal breaker for me, but I certainly did miss my physical lab notebook sometimes.

    Another problem I encountered was that MediaWiki was too cumbersome in terms of uploading things like equations, pictures of gels, diagrams, and so on; that made me tend to either procrastinate in updating my notebook, or just keep a de facto offline notebook part of the time. Those problems could probably be solved by picking my notebook format and software more carefully.


  2. Axel    16 March 08    #

    I run a biotech lab at an undergrad institution and usually have a couple of students in the lab at the same time. We are also working on three different projects at the same time. Therefore, it is usually hard for me to keep track of what is going on in the lab. I found it most useful to let students write an online journal summarizing progress in the lab. This does not replace a notebook – instead it complements it. Whereas the notebook contains all the experimental details, the journal just reflects what worked and what did not. Thus both, the students and I, can keep track of our results much more easily. It also allows me to see what the students struggle with and I can read it a night from home. On a busy day (ususally 7 days a week), this turns out to be a real advantage. The only problem is still that students don’t like to write :-)


  3. Frederick Ross    17 March 08    #

    I’ve been using an electronic notebook for a while now. I use the labbook class in LaTeX, organized by date then experiment, and I keep everything from an experiment under the same date instead of breaking it across days.

    For computer files, I have a directory support_files with subdirectories for year then month, then one for each experiment, where I keep all the necessary files, both images to be embedded in the lab book, code I’ve run, files, etc.

    I print it out and put it in a binder with page protectors, keep a backup copy on an external hard drive and the server the IT department here provides to the lab.

    Actually, I keep a couple notebooks in parallel: one for cloning, one for experiments, one for theory and analysis. This has worked out reasonably well.

    I don’t worry about being able to demonstrate that I did what I said when I said. I’m in a relatively slow and noncompetitive field. But if I were in a field where I had to be able to back up what I’m doing with hard records, I would keep my lab book in git (http://git.or.cz/), which uses a SHA1 hash to locate commits. Commit daily, keep a hard record of the SHA1 strings from the latest commit, and then you can guarantee that the history is what it appears to be all the way back to the beginning. This is something you might keep in a hard-bound journal with numbered pages, just writing line after line of 20 digit hashes.


  4. guy barbato    17 March 08    #

    i am a genetics faculty member and have a small lab. we’ve been using pacific.northwest.national.labs eln for several years:
    http://collaboratory.emsl.pnl.gov/software/eln/

    it works on any platform (linux/mac/windows) and is served via apache/tomcat. simple to set up; easy to use; trivial to backup/save.

    it is no longer being developed… so, we don’t have the level of integration that i’d like to see (e.g., with genomic databases, etc). but the code is available.

    i highly recommend it on several levels… but, your comment on keeping parallel e- and paper- notebooks is well taken.


  5. Damien Marsic    17 March 08    #

    I have been using ORNL enote1.12 for several years. It’s a web based open source electronic lab notebook. Easy to configure, adapt, backup and use. It’s written in Perl but a new Java version is coming soon.

    http://www.csm.ornl.gov/~geist/java/applets/enote/#demo

    I installed it on a local Linux server in our lab and restricted access to selected users based on IP address. I configured it so that each user can view (but not modify) everyone else’s notebook. It’s easy to do text searches and to include pictures or any other file.

    Damien


  6. Lee Tanzer    17 March 08    #

    Two points:

    1. What are the provisions for signature and counter-signature in your electronic notebooks to satisfy patent protection requirements.

    2. Have any of you tried Evernote, http://www.evernote.com/products/ software – basically a endless electronic paper tape to capture notes, images, documents,web pages…
    It has OCR, hand writing recognition, and electronic forms capabilities, as well as auto and manual indexing capabilities


  7. PhilipJ    17 March 08    #

    Great comments all, keep them coming!

    Lee, the signature issue did come up, and in the commercial software there is apparently a way to apply digital signatures. With a private wiki or what have you, this is less obviously possible, but someone did bring up the idea of printing things out, and sending them to yourself via regular snail mail, with your signature across the back, etc, and leaving it closed until such a case that it might be required. I’m not really in that business though, so I’m not sure if that’s sufficient for patents.


  8. Patrick    18 March 08    #

    CambridgeSoft’s Desktop and Workgroup E-Notebook is a popular choice among academic and small company settings.

    10 of the top 15 pharmaceutical companies use the Enterprise E-Notebook for IP storage, workflow enhancement and scientific collaboration.

    over 300,000 people total have access to the CambridgeSoft E-Notebook.

    www.cambridgesoft.com


  9. john    18 March 08    #

    We tried Cambridgesoft’s notebook last year, and it was a horrible experience – clunky and like something out of the 90s. We also discovered that the data we’d put into the desktop version couldn’t be moved up to the Enterprise version without us paying through the nose. Plus one of my former supervisors who’s now at one of those pharmaceutical companies says that they’re having to pay for all their custom work to be done again when they upgrade the system. I’d steer clear and stick to paper for the moment.


  10. Jean-Claude Bradley    23 March 08    #

    My organic chemistry group has used the free hosted wiki service at Wikispaces for a few years now and I think it replaces paper fairly well. With the amount of raw data generated in our field, it is not too difficult to link to all of it from the wiki.

    But a wiki is not a good vehicle for communicating milestones and explaining the big picture of our research – we use a blog for that and link back to the relevant experiments on the wiki.


  11. Alexander Polonsky    24 April 08    #

    There are many labs that are happily using iPad, a free, flexible yet structured ELN: http://ipadeln.com. Give it a try as it’s both simple and powerful.


  12. Fred    29 April 08    #

    I’ve been using Journler [ http://www.journler.com/ ].


Name
Email
http://
Message
  Textile Help