Data Analysis with R Workshop

Software

You are expected to bring your own computer!

Before the workshop, please install:

1. R (v3.5 or later), and
2. RStudio Desktop (v1.1 or later)

Both packages are cross-platform and free.

At the beginning of the workshop we’ll initialize an RStudio project that I have already prepared for you. Thus you should have the software installed and a functional internet connection. Please confirm with you coordinator if you need special access privileges where the workshop is being held.

Appropriate Data Sources

I want everyone to leave the workshop with the skills and confidence to begin tackling their own data analysis needs using R!

Unless explicitly stated, this means that you are expected to bring in a dataset that you want to work on!

In previous workshops, participants have brought in a wide variety of data sources, so I’m looking forward to a challenging and lively workshop.

To keep it simple, a flat text file or Excel file will be the easiest thing to start with.

Think about your data analysis objective. What is your research question? What do you want to find out from your data? I’m not going to tell you what to do, but I’ll do my best to tell you how. Thus, the better you can articulate the what, the easier it will be for me to help you with the how.

Inappropriate Data Sources

Special data types, such as

• Images,
• Sequencing,
• Flow cytometry, or
• Mass spectrometry reads,

are beyond the scope of this workshop.

Is it possible to work with these kinds of data in R, but you’ll probably have to install additional packages to read in and analyze them (e.g. from BioConductor, which is not covered in this workshop). You’re likely to encounter idiosyncratic data structures which can be pretty difficult for beginners to wrap their heads around. This makes for an overly-complex and discouraging first encounter with R, which is why I prefer to begin with flat text files.

If this is what you’re working on, then it’s preferable to bring in extracted values instead, e.g. output from ImageJ or MaxQuant, that we can begin processing immediately in R. Then you can expand your workflow to encompass more of your data analysis pipeline afterwards.

Day 1

Day 2

Day 3