Project management with R
Workshop for scientists
As researchers grow in their skills, eventually the infrastructure surrounding the analysis will become an important catalyst for the reproducibility and longevity of analysis artefacts. Researchers are capable of creating complex and impressive projects, but it is unlikely to have benefited from any formal training or mentorship related to topics regarding project management. As a result of this, reproducibility and longevity of projects are likely to suffer. Learning aspects of project management is a vital, but rarely taught, skill when working on a project. While project management has some general guidelines that traverse fields, there are specific aspects to think about when working R, given R’s idiosyncrasies.
Understand what a working environment is and how R deals with them
Successfully implement a project-oriented workflow
Understand best-practice naming of files
Know how to practice safe paths from R
Understand basics of R package management
Schedule 1
Saving source and blank slates | 1 hour |
Project-oriented workflow | 1 hour |
How to name files & practicing safe paths | 1 hour |
Lunch break | 30 minutes |
R package management | 1.5 hours |
Preparations
You might already have R and RStudio installed on your system. We highly recommend updating both of them to the latest version before the workshop, to ensure you can follow along all the exercises.
A new version of RStudio and R is recommended. Even if you have R and RStudio installed on your system, you should update all them before the workshop if you installed them longer than 2 months ago.
R packages
In addition to R and RStudio, a series of R packages will also be needed to complete the workshop tasks.
# Run in R
install.packages(c(
"remotes", # installing packages from GitHub
"rmarkdown", # rendering reports
"fs", # file system operations
"here", # navigating paths
"usethis", # for course materials
"tidyverse", # data-wrangling
"renv" # package management
))
If you are using a windows computer, you will also need to install RTools:
https://cran.r-project.org/bin/windows/Rtools/
RTools is an important addition to the R ecosystem.
Most institutional software handlers will have RTools as an option to install.
Resources
Footnotes
This course relies heavily on materials from What They Forgot to Teach you about R, which is an excellent resource to dig deeper into to improve the way you work in R.↩︎