About me

• PhD in statistics from Iowa State (defended in December)

  • Taught intro to business statistics for 4 years.
  • Thesis: Interfacing R with Web Technologies for Interactive Statistical Graphics and Computing with Data

• Maintainer of plotly's R package (for nearly 2 years!)

  • Authored the plotly book in 2016
  • Received the John M. Chambers Statistical Software Award in 2017

• Author of many other R packages

  • LDAvis, animint, pitchRx, rdom, etc

• Delivered a handful of R workshops

  • Most recently (solely) delivered an R workshop at plotcon

About this workshop

• Mostly a re-hash of my plotcon workshop

• Day 1: Getting (re)-acquainted with R, RStudio, data wrangling, ggplot2, and plotly

  • Basic R syntax and data structures
  • Using ggplotly() to convert ggplot2 to plotly
  • Using layout()/style() to modify plotly objects
  • Tips and info for getting "un-stuck"

• Day 2: Advanced plotly

  • Animation
  • Highlighting/filtering in multiple linked views
  • Capturing plotly events in shiny

library(plotly)
usa <- albersusa::usa_sf("laea")
p <- ggplot(usa) + geom_sf(aes(fill = pop_2010)) 
ggplotly(p)

Various ways to customize

p <- ggplot(usa) + 
  geom_sf(aes(fill = pop_2010, text = paste(name, "had", pop_2010, "\n in 2010"))) 
style(ggplotly(p, tooltip = "text"), hoverlabel = list(bgcolor = "white"))

Define custom interactions

The code

library(crosstalk)
usd <- SharedData$new(usa)
p <- ggplot(usd) + geom_sf(aes(fill = pop_2010))
highlight(
  ggplotly(p), "plotly_hover", opacityDim = 0.8,
  selected = attrs_selected(line = list(color = "black"))
)

Annotate on click

Highlighting in small multiples

d <- subset(txhousing, city %in% c("Galveston", "Midland", "Odessa", "South Padre Island"))
sd <- SharedData$new(d, ~year)
p <- ggplot(sd, aes(month, median, group = year)) + geom_line() +
  facet_wrap(~city, ncol = 2)
gg <- ggplotly(p, tooltip = "year")
highlight(gg, on = "plotly_click")

Making comparisons with dynamic brush

highlight(
  gg, on = "plotly_click", dynamic = TRUE, persistent = TRUE, selectize = TRUE
)

Animation

p <- ggplot(sd, aes(month, median)) + 
  geom_line(aes(group = year), alpha = 0.2) +
  geom_line(aes(frame = year), color = "red") +
  facet_wrap(~city, ncol = 2)
ggplotly(p) %>% animation_opts(1000)

Filter

s <- filter_select("years", "Years", sd, ~year)
p <- ggplot(sd, aes(month, median, group = year)) + geom_line() + facet_wrap(~city, ncol = 2, scales = "free")
htmltools::tagList(s, ggplotly(p, dynamicTicks = T))

Summary

• The ggplotly() function makes it easy to create easy interactive graphics via ggplot2.
• Any plotly R object may be modified/customized via the plotly.js figure reference.
• Crosstalk's SharedData class may be leveraged to:
  • Pose queries (via indirect/indirect manipulation)
  • Make comparisons (via persistent/dynamic highlighting)