Project management/notebook apps for bioinformaticians

At work recently, we’ve been discussing electronic notebooks, project management tools and so on. Here’s the problem: like other researchers, computational biologists would like to keep some record of what they’ve done during their working day. This is necessary so as to keep track of your progress, to reassure your boss that you really are working and possibly (though rarely), if legal issues like IP ever come up. Of course people like us refuse to maintain a paper notebook, so what do we do?

My suggestion that my directory hierarchy and filesystem time stamps serve perfectly well as a measure of progress was understandably not well received :). Another colleague pointed out that if your main focus is writing code, a CVS repository serves as an indicator of your progress. Then there are various blogging apps that you might adopt to suit such as Alf’s NotePress. Some Googling for “electronic lab notebook” throws up a few projects, most of which are either long dead or unsuitable. Similarly “project management software” yields a lot of stuff more suited to a corporate environment with budgets, tickets and so on.

What we need is some sort of integrated system with these features:

  • Web-based (probably PHP/MySQL)
  • Notebook/diary to record what you did in a day, plans, thoughts
  • Access to files that you generated (text, graphics, plots, code), perhaps via a CVS repository
  • Literature management? Bookmark references relevant to your project or something more complex, like RefDB integration
  • Documentation, wiki-like
  • Possibly multi-user, space for discussion, comments and so on

There are plenty of tools that accomplish each of these tasks separately, but I wonder if anyone out there is working on an integrated system? The thought of wrapping all these things up in some kind of CMS appeals to me but I don’t know if I can spare the time.

2 thoughts on “Project management/notebook apps for bioinformaticians

  1. The age-old problem… My current solution is to set up a notepress install on our internal-only web servers to act as electronic labbooks, and maybe patch in standard RSS-based P/Hubmed queries. Much less than optimal as a solution, though.

    I suspect glueing all the features you list is fairly trivial, wteo the bib management. I’ve given up on RefDB multiple times, due to truly annoying lib bleeding-edge dependecies, with which I don’t want to break my system. I really don’t see an alternative, though.

    btw, a neat trick I’ve been using for a while is to index a local pdf collection with swish-e, for full text searching. Trivial to implement, but very powerful.

    Here’s a wild and wacky thought – let’s put this up as a nodalpoint project, and maybe actually work on the damned thing…

  2. I’ve been meaning to look at your swish-e solution for a while – maybe now is the time. Another indexing option that I once considered was a local ht-dig, which might integrate with the web app idea. And I like the idea of a nodal project.

    See next post for my refdb frustrations. Really annoying because it used to work just fine, but I never got it running as a tool for daily use. When you suddenly can’t even compile it anymore it’s tempting to assume that you’re doing something wrong. Admittedly Debian unstable has gone to MySQL 5 which has affected a few apps of late, but not to breaking point. The refdb maintainer also seems very involved with libdbi which explains his cutting edge predilection I guess.

Comments are closed.