I became interested in genomics towards the end of my time as an undergraduate. Experiments in genomics generally result in the production of incredibly large data-sets that can only really be manipulated and analysed using computational methods. Through a combination of classes, self-teaching and internet forums I have managed to become moderately competent at R, perl and shell scripting. It also really helped to be surrounded by a number of excellent computer scientists during graduate school for those moments when I realized I was above my head.

• Statistics, you can never ever ever ever have enough statistical knowledge. The more you acquire the more you realize you need.
• Patience. So much of genomics involves careful planning to produce a single, well thought out data-set. This data-set can then take months or years to analyze
• Forethought. As outlined above, you have to consider your experimental design before you begin collecting genomics-scale datasets
• Collaborative abilities. As the field of genomics and biology becomes more and more complex there comes a point where you are probably going to need outside help. Having the ability to reach outside your field often comes with the need for a unique set of skills.
• The ability to admit you have reached your limit. Following on from the ability to collaborate is the ability to recognize when you need to start a collaboration. At some stage you will hit the limit of your abilities as a programmer. You need to admit that you have reached that limit and look out for someone that can help bridge the gap.

Mostly it was through necessity. After a short class on perl and R I have had to teach myself many of the skills that I use on a regular basis. Some of this teaching came from books like the O'Reilly series, but the internet has an extensive collection of forums like stackoverflow etc. that can be searched. I think generally I have started with an idea of a test or plot I want to generate and crawled through a variety of these forums until I eventually figured out how to do what I want. It's not always pretty, but has gotten 99% of what I need to do done.

The dynamics of transcription and how it is regulated on a genome-wide basis