Building a shiny app

SISBID 2026
https://github.com/dicook/SISBID

Shiny Apps in R

• shiny package developed and supported by RStudio (Winston Chang, Joe Cheng, JJ Allaire and others)
• Resources for shiny: https://shiny.rstudio.com/:
  • Gallery of great examples,
  • Tutorials at different levels

Are these notes genuine?

Swiss banknote data (Flury, B. and Riedwyl, H. 1988): measurements on 100 genuine and 100 counterfeit old-Swiss 1000-franc bank notes

Status: genuine or counterfeit

Can we classify the status visually?

Visual classification

banknote <- readr::read_csv(here::here("data/banknote.csv"))

banknote |> 
  ggplot(aes(x = Right, fill = Status)) + 
  geom_histogram(binwidth = .1)

Counterfeits have on average a wider right edge. What about the other measurements?

A first shiny app with ggpaintr

Make a shiny app to explore histograms:

{banknote |> 
  ggplot(aes(x = ppVar(Right), fill = Status)) + 
  geom_histogram(binwidth = .1) } |> ggpaintr::ptr_app()

ggpaintr keywords: ppVar (make a variable widget), ppNum (make a number widget)

check the Addins > ggpaintr placeholders

You do it

1. Run the ggpaintr code in your session.

2. Determine: which measurement does distinguishes the best between counterfeits and genuine banknotes

3. Rewrite the expression to show a scatterplot of two measurements. Don’t forget to make the y-axis interactive as well!

4. Which combination of variables allows a perfect distinction?

How does the magic work?

Create your second app

• In RStudio, File menu, New file, Shiny web app to start a new app
• The easiest start is the Single file, which will put both server and ui functions in the same file, app.R
• Make sure to change the number of bins!

Two main parts

• What we see and interact with:
  • user interface: layout with user input and (plot) output

• What is going on underneath:
  • the server: glue between user input and output

A Minimal Example

library(shiny)

ui <- fluidPage(
)

server <- function(input, output, session) {
}

shinyApp(ui, server)

A bit more fancy

library(shiny)
sidebar <-  sidebarPanel(width = 3, "Fun inputs")

main_col <- column(width = 9, "Some results")

ui <- fluidPage(
  title = "App Title",
  sidebar, 
  main_col
)

server <- function(input, output, session) {
}

shinyApp(ui, server)

A bit more fancy

library(shiny)
sidebar <-  sidebarPanel(
  width = 3, 
  textInput("name", "Enter your name:", value = "Susan")
)

main_col <- column(width = 9, "Some results")

ui <- fluidPage(
  title = "App Title",
  sidebar, 
  main_col
)

server <- function(input, output, session) {
}

shinyApp(ui, server)

Shiny Inputs

Shiny has many different input options, see the widget gallery:

• actionButton() - creates a clickable button
• selectInput() create a select list
• checkboxInput() and checkboxGroupInput()
• dateInput() - calendar to select a date
• dateRangeInput() - select a range of dates
• fileInput() - upload a file
• numericInput() - input a numeric value
• radioButtons() - select one or more items
• sliderInput() - slide along a range of values
• textInput() - input a string

Your Turn

Add a list of your favorite countries to the sidebar panel:

library(shiny)

sidebar <-  sidebarPanel(
  width = 3, 
  textInput("name", "Enter your name:", value = "Susan")
)

main_col <- column(width = 9, "Some results")

ui <- fluidPage(
  title = "App Title",
  sidebar, 
  main_col
)

server <- function(input, output, session) {
}

shinyApp(ui, server)

Adding Output

• Output needs to be specified in two places:
  • in the ui function we need to create a placeholder to reserve space for the output
  • in the server function, we need to create the output.

1. ui components are usually xxxOutput (outputID)
   Output can be in form of tables, plots, text, …

2. server components are renderXXX, where xxx is a table, plot, text, etc.

Shiny Outputs

Shiny also has a variety of different output options:

• renderDataTable() - outputs an interactive, sortable data table
• htmlOutput() - output html elements
• renderPlot() - output an R plot
• renderPrint() - output text from print() in R
• renderTable() - output an HTML table
• renderText() - output text from R
• renderUI() - output a custom part of the user interface
• renderImage() - print an image to the page

Adding a plot

library(shiny)
library(ggplot2)
library(dplyr)

sidebar <-  sidebarPanel(
  width = 3, 
  selectInput(
    "country", "Pick your favorite country:", 
              choices = c("Australia", "Brazil", "China")
  )
)

main_col <- column(                               #<<
  width = 9,                                      #<<
  plotOutput("scatter")                           #<<
)                                                 #<<

ui <- fluidPage(
  title = "App Title", sidebar, main_col
)

server <- function(input, output, session) {
  output$scatter <- renderPlot({                  #<<
    mtcars %>% ggplot(aes(x = disp, y = mpg)) +   #<<
      geom_point() +                              #<<
      ggtitle(input$country)                      #<<
  })                                              #<<
}

shinyApp(ui, server)

Your turn

• Add an interactive plot of the TB data into your app.
• Change the select box, so that the user can choose any country. Think of unique(tb$country) for defining the choices

Deploy an app

• Sign up for an account on https://www.shinyapps.io/
• Authenticate your account
• You may need to do some setup in your session, e.g. install the library rsconnect

Resources

• RStudio Tutorial
  
• Deploy your app for others to use
• Shiny Setup, Showcase, and Server
  
• Community discussion

This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.