Scientists generate data, and with modern methods a lot. For some time, spreadsheets were enough for decent analysis and visualisation, offering simple formulas, templates and production of the figures. However, with the increase of file size and generally larger number of measurements, many researchers start to fees that this is just not enough. Suddenly we are in the position that we measure much more than we can process. For example, in our case when we measure with PTR-MS, we generate around 2 GB of data in a day. Luckily, the projects usually last for a week or two.
Now the question is: at what point investing the time in learning a scripting language pays off? Actually, you may be surprised how soon that happens. Scripting has power in data manipulation and plotting, but the time-saver is also code recycling. Ones you have a good script you recycle it a lot, and what you reuse is not just code lines but concepts, you know what works and what doesn’t for your data.
Now, the question is where to start with data programming, and my answer is R!
Here are the reasons why: (1) R is an excellent platform for data importing, statistics, and printing the plots - a lot’s of figures in no time; (2) in just couple of lines you achieve a lot; (3) you can execute fragment by fragment of code, which makes debugging easy; and (4) almost everything is already done by someone, and the concept is available online, so you just need to copy the code from the net and modify it.
Time paradox !
Learning and using a programming language has its time paradox. You do invest a lot of time, but you never feel it, and it never gets boring. So when you start, prepare yourself to learn to stop!