A new direction for leesjohn

Since October 2013 I have stopped using Fedora, and instead use machines running Ubuntu 12.04/13.10, Windows 8 and OS X 10.8.5. As these OSs have a larger user base than Fedora, many of the issues I encounter are well documented and easy to fix (i.e. there is a stackexchange post as one of the top three google results), hence there haven’t been many things for me to post under the original remit of this blog.

Of course, when I do encounter an undocumented OS based issue as I go about my business I’ll still try and post it on leesjohn. However I expect this to be much less common than previously, and the new computing based issues I find myself having to deal with are:

• Interactions and differences between OS X and Ubuntu when working with them simultaneously
• Working with Ubuntu without a sudo account (e.g. installing software, using custom libraries)
• Use of radio software (e.g. Rivendell, Cuedex, Jack)

I have now changed area from physics to bioinformatics, and think there is scope to share many of the scripts and programs I write for this, as well as solutions to issues I encounter in the area. So I have finally gotten round to setting up a github account (https://github.com/johnlees) to share as much of the code I write as possible.

From now on leesjohn will primarily be to document the scripts in these repositories, and to share some original tools. I’ve already committed some things, which you can see at:

• https://github.com/johnlees/bioinformatics (for bioinformatics tools)
• https://github.com/johnlees/config (for software related configuration)
  Hopefully this will make it very easy if someone ever does want to use some of the stuff I’ve written.

In the next few weeks I am hoping to write some posts about some of the more useful/general things in these repositories. I am also planning on making a wrapper script to allow you to use impute2 (http://mathgen.stats.ox.ac.uk/impute/impute_v2.html) to infer your whole genome from the ‘raw data’ you get if you have had a 23andme done (https://www.23andme.com/) - which as far as I can tell is not something yet available in the public sphere, but something I think many of 23andme’s clients could be interested in.