Clean Coding & Project Organization

Running R from Command Line

Clean coding

The main goal is readability

"You're always working on a project with at least 1 other collaborator,
and that is future you" Hadley Wickham

Anyone should be able to pick up the code and understand what is does.

Use

• descriptive names
  
• consistent style (indentation, names)
  
• functions (single operation, single function)
  
• tell us what the function is doing, not how

Be consistent

 
see e.g. Google Stylesheet

Use functions

Break down the problem into functions


One operation, one function

Comments

"Write code that is so clear that you don't need to comment it, and then
comment it anyway." Tracy Teal

• Write things that will make your life easier when you come back to a project:

• Avoid writing comments that just repeat what the code says

Some best practises

Different styles are possible. These are just some suggestions (learned the hard way)

Keep all your source files for a project in the same directory.

Then use relative paths.

df <- read.table("files/data.csv", header = TRUE)

rather than

df <- read.table("/home/wyder/PROJECTS/coexpression/files/data.csv", header = TRUE)

More tips

• Run R inside of the project folder
  
• Avoid setwd() in a script. Never use setwd() once the script is running. If you use it put it at the very beginning.
  
• Don't save a session on exit. Prevent unexpected results and start with a clean enviroment
  
• Store the versions of R/packages you used with sessionInfo()
  
• Write down project goals (README file)
  
• Use a versioning system (e.g. git)

For larger projects separate your scripts

Bioconductor

www.bioconductor.org

Tools for the statistical analysis and comprehension of genomic data in the R programming language.
Also data integration.

• Sequence Analysis
  
• RNA-seq
  
• Oligonucleotide Arrays
  
• Variants
  
• Annotation (Gene, Pathway, GO, …)
  
• Mass spectrometry and proteomics
  
• …

Sources

Software Carpentry