R/cohort_dataset_selection_server.R

In CRI-iAtlas/iatlas.modules2: iAatlas Modules2

cohort_dataset_selection_server <- function(
  id,
  default_datasets = shiny::reactive("TCGA"),
  dataset_type     = shiny::reactive("analysis"),
  display_module_availibility_string = shiny::reactive(T)
) {
  shiny::moduleServer(
    id,
    function(input, output, session) {
      ns <- session$ns

      choices <- shiny::reactive({
        options <- iatlasGraphQLClient::query_datasets(types = dataset_type()) %>%
          dplyr::select("display", "name") %>%
          tibble::deframe(.)
       
        if(dataset_type() %in% c("analysis", "scrna")) return(options)
        else return(create_ici_options(options)) #for the ICI Cohort Selection, we have RNA-Seq and Nanostring data. This function returns a list organizing them. 
      })

      output$dataset_selection_ui <- shiny::renderUI({
        shiny::req(choices(), default_datasets())
          
        label <- shiny::reactive(
            switch (
             dataset_type(),
             "analysis" = "RNA-seq data",
             "ici" = "RNA-seq data",
             "scrna" = "single-cell RNA-seq data"
            )
        )
        
        if(dataset_type() == "analysis"){
          shiny::selectInput(
            inputId  = ns("dataset_choices"),
            label    = "Select Datasets",
            choices  = choices(),
            selected = default_datasets()
          )
        }else if(dataset_type() == "scrna"){
            shiny::checkboxGroupInput(
                inputId  = ns("dataset_choices"),
                label    = "Select Datasets",
                choices  = choices(), #list scRNA-seq datasets
                selected = default_datasets()
            )
        }else{ #ICI Cohort selection will have two dataset columns, one for RNA-Seq and the other for Nanostring datasets
          shiny::fluidRow(
            shiny::column(
              width = 6,
              shiny::checkboxGroupInput(
                inputId  = ns("dataset_choices"),
                label    = paste0("Select Datasets - ", label()),
                choices  = choices()[[1]], #list datasets
                selected = default_datasets()
              )
            ),
            if(dataset_type() == "ici"){ #ICI Cohort selection will have two dataset columns, one for RNA-Seq and the other for Nanostring datasets
                shiny::column(
                  width = 6,
                  shiny::checkboxGroupInput(
                    inputId  = ns("dataset_choices_ns"),
                    label    = "Select Datasets - Nanostring data (only Immunomodulators module)",
                    choices  = choices()[[2]], #list nanostring datasets
                    selected = default_datasets()
                  )
                )
            }
          )
        }
      })
      
      dataset_selection <- shiny::reactive({
        if(dataset_type() %in% c("analysis", "scrna")) return(input$dataset_choices)
        else return(c(input$dataset_choices, input$dataset_choices_ns)) #ICI Cohort selection can have RNA-Seq and Nanostring ds selected
      })

      # This is so that the conditional panel can see the various shiny::reactives
      output$display_module_availibility_string <- shiny::reactive({
        display_module_availibility_string()
      })

      shiny::outputOptions(
        output,
        "display_module_availibility_string",
        suspendWhenHidden = FALSE
      )

      output$module_availibility_string <- shiny::renderText({
        shiny::req(input$dataset_choices)
        create_cohort_module_string(
          input$dataset_choices
        )
      })

      return(shiny::reactive(sort(dataset_selection())))
    }
  )
}