app.R

library(shiny)
library(shinydashboard)
library(shinyjs)
library(prophets)

ui <- dashboardPage(
  skin = "black",
  
  dashboardHeader(
    title = div(tags$img(src='hexlogo.png', height = '40', width = '40'), "PROPHETS")
  ),
  
  dashboardSidebar(
    
    useShinyjs(),
    
    sidebarMenu(id = "sidebar",
                tags$head(tags$style(".inactiveLink {
                                     pointer-events: none;
                                     cursor: none;
                                     }")),
                menuItem("Home", tabName = "home", icon = icon("house")),
                menuItem("PFS ratio", tabName = "pfsratio", startExpanded = TRUE,
                         menuSubItem("Data upload", tabName = "dataupload", icon = icon("upload")),
                         menuSubItem("PFSr analysis", tabName = "pfsra", icon = icon("subscript")),
                         menuSubItem("PFSr visualization", tabName = "pfsrv", icon = icon("chart-column"))
                ),
                menuItem("Sample size calculator", tabName = "scalc", icon = icon("calculator"))
    )
    
  ),
  
  dashboardBody(
    
    tabItems(
      
      tabItem(tabName = "home",
              h1("PROPHETS: PROgression-free survival ratio as primary endpoint in PHasE 2 TrialS in oncology"),
              
              br(),
              
              h3("Welcome to the PROPHETS shiny app, an online modelling and prediction tool designed to help clinicians.
           
           PROPHETS is an R-Package which is developed to collect and implement different methods for the design and analysis of trials that use PFSratio as their primary endpoint. 
           It provides a convenient wrapper around existing methods to calculate and plot PFSratio-based results. You can find the R package at https://github.com/huellejn/prophets"
              ),
              
              br(),
              
              h3("This shiny app is therefore meant as a user-friendly web calculator for clinicians and statisticians that aim to explore PROPHETS without the burden of R-based coding."
              ),
              
              br(),
              
              h3("The development of this software was a collaborative project between Fondazione IRCCS Istituto Nazionale dei Tumori of Milan, Italy, 
           and the German Cancer Research Center (DKFZ) of Heildeberg, Germany."),
              
              br(),
              
              h3("We welcome any feedback you may have about PROPHETS. 
        If you have questions about its development
           or there are features you would like to have added please let us know by emailing us at federico.nichetti@istitutotumori.mi.it .
                 Thanks!"),
              
              br(),
              
              h3(HTML(paste0("Federico Nichetti", tags$sup("1,2"), "& Jennifer Hüllein", tags$sup("2")))),
              h4(HTML(paste0(tags$sup("1"), "Medical Oncology Department, Fondazione IRCCS Istituto Nazionale dei Tumori di Milano, Milan, Italy"))),
              h4(HTML(paste0(tags$sup("2"), "Computational Oncology, Molecular Precision Oncology Program, 
                 National Center for Tumor Diseases (NCT) and German Cancer Research Center (DKFZ)")))
      ),
      
      tabItem(tabName = "dataupload",
              fluidRow(
                column(4, radioButtons("btn_delim", "Delimiter", choices = c("comma", "semicolon", "space", "tab"))),
                column(4,
                       fluidRow(
                         column(3, 
                                div(style = "display: inline-block; vertical-align: -20px;",
                                    checkboxInput("btn_header", "Header", value = TRUE))
                         )
                         
                       ),
                       fluidRow(
                         column(3, checkboxInput("btn_quotes", "Quotes", value = TRUE))
                       )
                )
              ),
              fileInput("file", label = NULL),
              DT::dataTableOutput("inputdata")
      ),
      
      tabItem(tabName = "pfsra",
              fluidRow(
                column(3, numericInput("delta", "Delta", min = 0.1, max = 3, value = 1, step = 0.1) ),
                column(3, numericInput("prob", "Probability", min = 0.01, max = 1, step = 0.01, value = 0.5) )
              ),
              fluidRow(
                column(3, checkboxInput("modified", "Modified", FALSE)),
                column(6, numericInput("min_pfs2", "Minimal PFS2 in months", min = 0, value = 0.6))
              ),
              fluidRow(column(12,
                              h4("Model summary"),
                              tableOutput("stats_summary"))
              )
      ),
      
      
      tabItem(tabName = "pfsrv",
              
              numericInput("delta_pfsrv", "Delta", min = 0.1, max = 3, value = 1, step = 0.1),
              fluidRow(
                column(3, checkboxInput("modified_pfsrv", "Modified", FALSE)),
                column(6, numericInput("min_pfs2_pfsrv", "Minimal PFS2 in months", min = 0, value = 0.6))
              ),
              tabsetPanel(
                tabPanel("Swimmer plot",
                         plotOutput("plot_swimmer")
                ),
                tabPanel("Correlation between PFS1 and PFS2",
                         checkboxInput("logscale", "log scale", value = FALSE),
                         plotOutput("plot_pfs_correlation")
                ),
                tabPanel("Cumulative hazard ratio",
                         checkboxGroupInput("cumhaz", label = NULL, choices = c("PFS1", "PFS2"), selected = c("PFS1", "PFS2")),
                         plotOutput("plot_cumhaz")
                ),
                tabPanel("Weibull plot",
                         plotOutput("plot_weibull")
                ),
              )
              
      ),
      
      tabItem(tabName = "scalc",
              fluidRow(
                column(6, 
                       radioButtons("method", "Method for sample size calculation", 
                                    choiceNames = c(
                                      "Sample size calculation", 
                                      "Sample size calculation using GES", 
                                      "Sample size calculation according to a pre-specified proportion of patients with positive PFSr", 
                                      "Calculation of the statistical power of a study with a given sample size",
                                      "Calculation of the statistical power of a study with a given sample size using GES"),
                                    choiceValues = c("scalc", "scalc_ges", "scalc_p", "pcalc", "pcalc_ges"),
                                    selected = "scalc"
                       )
                ),
                column(6, 
                       numericInput("alpha", "Alpha", 0.05, step = 0.05),
                       conditionalPanel(
                         condition = "input.method=='scalc' || input.method=='scalc_ges'",
                         numericInput("power", "Power", value  = 0.8, min = 0.5, max = 0.99, step = 0.01),
                       ),
                       conditionalPanel(
                         condition = "input.method=='scalc' || input.method=='scalc_p' || input.method=='pcalc'",
                         list(
                           numericInput("rho", "Rho", value = 0.5, min = -1, max = 1, step = 0.1),
                           numericInput("alt_hr", "Alternative PFS ratio", value = 1.33, min = 0.1, max = 3.0, step = 0.1)
                         )
                       ),
                       conditionalPanel(
                         condition = "input.method=='scalc' || input.method=='pcalc'",
                         list(
                           numericInput("null_hr", "Null PFS ratio", value = 1, min = 0.1, max = 3.0, step = 0.1),
                           numericInput("k", "k", value = 1, min = -1, max = 1, step = 0.1)
                         )
                       ),
                       conditionalPanel(
                         condition = "input.method=='scalc' || input.method=='scalc_ges' || input.method=='scalc_p'",
                         numericInput("lost", "Proportion of non-informative pairs (Lost)", value = 1, step = 0.1)
                       ),
                       conditionalPanel(
                         condition = "input.method=='scalc' || input.method=='pcalc'",
                         selectInput("model", "Model", choices = c("GBVE", "Weibull"), selected = "GBVE", multiple = FALSE)
                       ),
                       conditionalPanel(
                         condition = "input.method=='scalc_ges' || input.method=='pcalc_ges'",
                         numericInput("ges", "GES", value = 0.3, step = 0.1)
                       ),
                       conditionalPanel(
                         condition = "input.method=='scalc_p'",
                         list(
                           numericInput("p0", "P0 (Null hypothesis)", min = 0, max = 1, value = 0.15, step = 0.1),
                           numericInput("p1", "P1 (Alterantive hypothesis)", min = 0, max = 1, value = 0.24, step = 0.1),
                           numericInput("beta", "Beta", 0.1, step = 0.05)
                         )
                       ),
                       conditionalPanel(
                         condition = "input.method=='pcalc' || input.method=='pcalc_ges'",
                         numericInput("samplesize", "Sample size", min = 1, value = 150)
                       )
                )
              ), 
              fluidRow(
                column(12, tableOutput("scalc_results"))
              )
              
              
              #verbatimTextOutput("scalc_results")
      )
      
    )
    
  )
  
)



server <- function(input, output, session) {
  
  # Data upload
  
  ## Get the path to the file
  userFile <- reactive({
    # If no file is selected, don't do anything
    validate(need(input$file, message = FALSE))
    input$file
  })
  
  ## Read the data
  inputfile <- reactive({
    delim <- input$btn_delim
    read.delim(userFile()$datapath,
               header = input$btn_header,
               sep = ifelse(delim == "comma", ",",
                            ifelse(delim == "semicolon", ";",
                                   ifelse(delim == "space", "", "\t"))),
               quote = ifelse(input$btn_quotes == TRUE, "\"'", ""),
               stringsAsFactors = FALSE)
  })
  
  ## Print message if file was successfully uploaded
  observe({
    msg <- sprintf("File %s was uploaded", userFile()$name)
    cat(msg, "\n")
  })
  
  
  ## Check that necessary columns are present
  observeEvent(inputfile(), {
    
    columns_required <- c("PFS1", "PFS2")
    columns_actual <- names(inputfile())
    
    duration <- 5
    
    if( all(columns_required %in% columns_actual) ) {
      msg_dataUpload <- "All required columns are present."
      type <- "default"
      removeCssClass(selector = "a[data-value='pfsra']", class = "inactiveLink")
      removeCssClass(selector = "a[data-value='pfsrv']", class = "inactiveLink")
      
    } else {
      msg_dataUpload <-  paste("The following columns are missing:", paste(columns_required[!columns_required %in% columns_actual], collapse = ", " ) )
      type <- "error"
      addCssClass(selector = "a[data-value='pfsra']", class = "inactiveLink")
      addCssClass(selector = "a[data-value='pfsrv']", class = "inactiveLink")
    }
    
    showNotification(
      msg_dataUpload,
      duration = duration,
      closeButton = TRUE,
      type = type
    )
  })
  
  output$inputdata <- DT::renderDataTable({inputfile()}, rownames = FALSE, options = list(scrollX = TRUE, pageLength = 10))
  
  # PFSr analysis
  output$stats_summary <- renderTable({
    df <- inputfile()
    df$ratio <- df$PFS2 / df$PFS1
    
    if(input$modified == TRUE) {
      df <- modify_PFS(df, delta = input$delta, min_pfs2 = input$min_pfs2 )
    }
    
    res <- prophets_summary(
      data = df,
      delta = input$delta
    )
  })
  
  # PFSr visualization
  
  ## Swimmer plot
  output$plot_swimmer <- renderPlot({
    df <- inputfile()
    df$ratio <- df$PFS2 / df$PFS1
    if(input$modified == TRUE) {
      df <- modify_PFS(df, delta = input$delta_pfsrv, min_pfs2 = input$min_pfs2_pfsrv )
    }
    res <- swimmerplot_PFSr(
      df, 
      delta = input$delta_pfsrv)
    return(res)
  })
  
  ## Correlation of PFS1 and PFS2
  output$plot_pfs_correlation <- renderPlot({
    df <- inputfile()
    df$ratio <- df$PFS2 / df$PFS1
    
    if(input$modified == TRUE) {
      df <- modify_PFS(df, delta = input$delta, min_pfs2 = input$min_pfs2 )
    }
    res <- plot_correlation_PFS(
      df, 
      delta = input$delta_pfsrv,
      log_scale = input$logscale)
    return(res)
    
  })
  
  ## Cumulative Hazard Ratio
  output$plot_cumhaz <- renderPlot({
    df <- inputfile()
    df$ratio <- df$PFS2 / df$PFS1
    if(input$modified == TRUE) {
      df <- modify_PFS(df, delta = input$delta_pfsrv, min_pfs2 = input$min_pfs2_pfsrv )
    }
    res <- plot_cumHaz(
      df,  
      selected_PFS = input$cumhaz 
    )
    return(res)
  })
  
  ## Weibull plot
  output$plot_weibull <- renderPlot({
    df <- inputfile()
    df$ratio <- df$PFS2 / df$PFS1
    if(input$modified == TRUE) {
      df <- modify_PFS(df, delta = input$delta_pfsrv, min_pfs2 = input$min_pfs2_pfsrv )
    }
    res <- plot_weibull(df)
    return(res)
    
  })
  
  # Sample size calculator
  output$scalc_results <- renderTable({
    
    # sreq(iv$is_valid())
    
    # Get input values
    val_method <- input$method
    
    val_alpha <- input$alpha
    val_power <- input$power
    val_rho <- input$rho
    val_alt_hr <- input$alt_hr
    val_null_hr <- input$null_hr
    val_k <- input$k
    val_model <- input$model
    val_lost <- input$lost
    val_ges <- input$ges
    val_p0 <- input$p0
    val_p1 <- input$p1
    val_beta <- input$beta
    val_sample_size <- input$samplesize
    
    if(val_method == "scalc") {
      res <- PFSr_samplesize(
        alpha = val_alpha,
        power = val_power,
        rho = val_rho,
        alt_HR = val_alt_hr,
        null_HR = val_null_hr,
        k = val_k,
        model = val_model,
        verbose = FALSE)
    } else if(val_method == "scalc_ges") {
      res <- ges_PFSr_samplesize(
        ges = val_ges, 
        alpha = val_alpha, 
        power = val_power, 
        lost = val_lost, 
        verbose = FALSE
      )
    } else if(val_method == "scalc_p") {
      res <- PFSr_samplesize_proportion(
        p0 = val_p0, 
        p1 = val_p1,
        alpha = val_alpha, 
        beta = val_beta,
        pfsratio = val_alt_hr, 
        lost = val_lost,
        verbose = FALSE
      )
    } else if(val_method=="pcalc") {
      res <- PFSr_power_calculate(
        sample_size = val_sample_size, 
        null_HR = val_null_hr,
        alt_HR = val_alt_hr, 
        rho = val_rho, 
        alpha = val_alpha,
        k = val_k, 
        model = val_model,
        verbose = FALSE)
    } else if(val_method == "pcalc_ges") {
      res <- PFSr_power_calculation_ges(
        sample_size = val_sample_size, 
        ges = val_ges, 
        alpha = val_alpha,
        verbose = FALSE
      )
    }
    
    
    names(res)[names(res)=="n"] <- "required sample size"
    
    return(res)
    
  })
  
  
}

shinyApp(ui, server)