README

Course Information

BIO708 Biostatistics

Instructor - Akira Terui

Suggested Repository Structure

Github repository should be organized so you can locate files easily. My suggest

• /code : may contain script files (.R).

• /data_raw : may contain raw data files. This data should not be edited and

• /data_fmt : may contain formatted or derived data (e.g., after data cleaning etc.).

I do NOT recommend creating sub-sub-directories under any of the above sub-directories. It will create issues when you call files in those deep directories, as you must specify sub-dir/sub-sub-dir/sub-sub-sub.... I would rather set file naming rules within each sub-directory so that files are sorted by file type (see File Name section).

File name

• Don’t

  • Don’t use white-space, comma, and period. Never, ever. Use underscore instead (_)

  • Avoid capital in file names.

• Do

  • Use pre-fix to group files automatically. For example, you may use figure_xxx.R for files generating figures, data_xxx.csv for data files, etc. This will help you organize files because (1) they are automatically grouped together within a sub-directory, and (2) you can call specific types of files using list.files() in R.

  • Be mindful about the hierarchy when naming files. For example, the prefix may begin from the largest grouping category (e.g., figure, table, analysis, format, etc), then specify specific tasks within each file group.