How do you manage your bioinformatics lab notebook?

[dsenalik]

The question is simply one of how do you record what you do when doing bioinformatic analyses? The classical paper laboratory notebook clearly can't adapt well to scripts and programs and the sometimes massive output files.

I currently keep track of what I do using simple (but perhaps primitive) HTML documents, since I can link to output files and images easily, and insert tables, or when necessary sub-pages. I don't need much formatting, so I like this for its simplicity (<ul> is my friend). And then I can share password-protected pages with others.

However, I feel that there might be better ways to do this, so what do other people use?

[Jean]

Blog/Wiki format is good for me. You can copy code snippets, link files, upload images, and make notes.

[Rocketknight]

Spreadsheets in Google Docs have been helpful for keeping a record of information that other people can view if they need to.

[kga1978]

I use DevonThink pro, which is fantastic but it only works for Mac

[Dario1984]

I use the Sweave package for R. You write a LaTeX document that has your descriptions and code. When you run it through R, it executes your code segments and saves the outputs to disk and places embedding code into the file. When you convert your LaTeX document to PDF, the output objects are embedded into the PDF. This is useful because you can be sure that the code generated the result files you are seeing, rather than re-visiting an analysis months later, and noticing you have a few versions of the same graph saved in your folder and not being able to remember which one you sent to the biologists, or why.

[Georgino]

I think blog is a kind of personal diary (although published to net) with commenting and trackback while wiki is a whiteboard on net. If a page is frequently updated, wiki is better since it's a collaboration tool (imagine to share a whiteboard with your colleagues). Wiki has link-system between pages so that it is suitable for knowledge management, I think.

http://www.cvconsultants.co.uk/

[arolfe]

Wiki pages are good, as others have mentioned.

I also try to leave projects with a README file and one or more shell scripts that run the whole analysis start to finish (data collection from standard locations through to reporting) so that I (1) have the exact record of what I ran and (2) can easily rerun if necessary. This can get messy if you're rerunning the same thing often, then you need some kind of version control for the scripts (RCS, git, etc) but version control for the data is hard if it's a big dataset.

[ucpete]

I just keep a Word document open all day. Any code I write gets cat'ed and pasted into the notebook. Any commands I run, with output (provided it's not like 10 pages worth) get pasted into the notebook. If I update a script, I show the diff between the old and new version. If I make a figure or a plot, I take a screen shot and paste it into the notebook. At the end of the day I print it and put it in a binder on my desk, but I also save to PDF and actually check it into SVN. One thing I do is change my PS1 variable to show the date and time so that when I copy and paste a command, it's inherently time stamped. Super useful, relatively simple, and I have the most detailed lab notebook of anyone I know and it requires minimal effort.

[sklages]

I am using an outliner for my personal documentation of what I am doing and what I have done. In my case http://www.notecasepro.com/

[dariober]

My 2p...

I'm quite happy with Redmine which supports wiki pages, svn repositories and you can share projects with others. For keeping code I use google code.

[adaptivegenome]

basecamp works for me

[loba17]

Evernote ...

I am working with Evernote (http://evernote.com/) ... I used DevonThink for a while but I was never really happy. With Evernote I have everything I need wherever I am ... give it a try!