Data Carpentry for Biologists

Combining Data Manipulations in dplyr

Notes

Setup

install.packages(c('dplyr', 'readr', 'tidyr'))
download.file("https://ndownloader.figshare.com/files/2292172",
  "surveys.csv")
download.file("https://ndownloader.figshare.com/files/3299474",
  "plots.csv")
download.file("https://ndownloader.figshare.com/files/3299483",
  "species.csv")

Introduction

  • Combine a series of data manipulation actions
  • Do each action in sequential order

Intermediate variables

  • Run a command
  • Store the output in a variable
  • Use that variable later in the code

  • Repeat

  • Obtain the data for only DS, with no null weights, sorted by year, with only the year and weight columns
ds_data <- filter(surveys, species_id == "DS")
ds_data_no_null_weight <- drop_na(ds_data, weight)
ds_data_by_year <- arrange(ds_data_no_null_weight, year)
ds_weight_by_year <- select(ds_data_by_year, year, weight)

Do Exercise 11.1, Portal Data Manipulation

Pipes

  • Intermediate variables can get cumbersome if there are lots of steps
  • |> (“pipe”) takes the output of one command and passes it as input to the next command
  • Want to take the mean of a vector
  • Normally we would run the mean function with the vector as the input:
x = c(1, 2, 3)
mean(x)
  • Instead we could pipe the vector into the function
x |> mean()
  • So x becomes the first argument in mean
  • If we want to add other arguments they get added to the function call
x = c(1, 2, 3, NA)
mean(x, na.rm = TRUE)
x |> mean(na.rm = TRUE)
  • Questions?
surveys |>
  filter(species_id == "DS")

ds_weight_by_year <- surveys |>
  filter(species_id == "DS") |>
  drop_na(weight)

ds_weight_by_year <- surveys |>
  filter(species_id == "DS") |>
  drop_na(weight) |>
  arrange(year) |>
  select(year, weight)

Do Exercise 12.1, Portal Data Manipulation Pipes.

The magrittr pipe

  • You will also see another type of pipe character %>%
  • This is the original pipe in R and you had to load the magrittr package to use it (this gets loaded automatically by dplyr)
  • |> is now considered the standard unless you know you need special magrittr functionality

Keyboard Shortcut

  • Shortcut: Ctrl-Shift-m