midasdashboard: MiDAS dashboard shiny app

#' The application server-side
#' 
#' @param input,output,session Internal parameters for {shiny}. 
#'     DO NOT REMOVE.
#' @import shiny
#' @import shinydashboard 
#' @import ampvis2
#' @import ggplot2
#' @importFrom plotly ggplotly renderPlotly layout
#' @importFrom tidyr spread
#' @importFrom data.table data.table fread melt
#' @importFrom dplyr arrange mutate bind_cols desc select filter
#' @importFrom magrittr %>%
#' @importFrom stringr str_remove_all
#' @importFrom DT renderDataTable
#' @importFrom openxlsx write.xlsx
#' @noRd
app_server <- function( input, output, session ) {
  ######################### DATA #########################
  data <- reactive({
    if(input$dataset == "MiDAS kvartalsprøver AS") {
      midas_plants <<- input$wwtp_midas
      d <- midasdashboard::MiDAS
      if(!is.null(input$wwtp_midas)) {
        ds <- suppressMessages(amp_subset_samples(d, Plant %in% midas_plants))
      } else
        ds <- d
    } else if(input$dataset == "MiDAS OnlineKontrol") {
      req(input$samplecontent_midasok)
      midasok_plants <<- input$wwtp_midasok
      d <- midasdashboard::midasok
      if(!"Alle" %in% input$samplecontent_midasok) {
        d <- suppressMessages(amp_subset_samples(d, SampleContent %in% input$samplecontent_midasok))
      }
      if(!is.null(input$wwtp_midasok)) {
        ds <- suppressMessages(amp_subset_samples(d, Plant %in% midasok_plants))
      } else
        ds <- d
    } else if(input$dataset == "MiDAD kvartalsprøver AD (bakterier)") {
      midad_db_plants <<- input$wwtp_midad_bac
      d <- midasdashboard::midad_bac
      if(!is.null(input$wwtp_midad_bac)) {
        ds <- suppressMessages(amp_subset_samples(d, Plant %in% midad_db_plants))
      } else
        ds <- d
    } else if(input$dataset == "MiDAD kvartalsprøver AD (archaea)") {
      midad_da_plants <<- input$wwtp_midad_arc
      d <- midasdashboard::midad_arc
      if(!is.null(input$wwtp_midad_arc)) {
        ds <- suppressMessages(amp_subset_samples(d, Plant %in% midad_da_plants))
      } else
        ds <- d
    } else if(input$dataset == "BioBank AD (bakterier)") {
      biobank_db_plants <<- input$wwtp_biobank_ad_bac
      d <- midasdashboard::biobank_ad_bac
      if(!is.null(input$wwtp_biobank_ad_bac)) {
        ds <- suppressMessages(amp_subset_samples(d, Plant %in% biobank_db_plants))
      } else
        ds <- d
    } else if(input$dataset == "BioBank AD (archaea)") {
      biobank_da_plants <<- input$wwtp_biobank_ad_arc
      d <- midasdashboard::biobank_ad_arc
      if(!is.null(input$wwtp_biobank_ad_arc)) {
        ds <- suppressMessages(amp_subset_samples(d, Plant %in% biobank_da_plants))
      } else
        ds <- d
    } else if(input$dataset == "BioBank AS") {
      biobank_as_plants <<- input$wwtp_biobank_all
      d <- midasdashboard::biobank_as
      if(!is.null(input$wwtp_biobank_all)) {
        ds <- suppressMessages(amp_subset_samples(d, Plant %in% biobank_as_plants))
      } else
        ds <- d
    } else if(input$dataset == "Sverige (bakterier)") {
      biobanksweden_bac_plants <<- input$wwtp_biobanksweden_bac_all
      d <- midasdashboard::biobanksweden_bac
      if(!is.null(input$wwtp_biobanksweden_bac_all)) {
        ds <- suppressMessages(amp_subset_samples(d, Plant %in% biobanksweden_bac_plants))
      } else
        ds <- d
    } else if(input$dataset == "Sverige (archaea)") {
      biobanksweden_arc_plants <<- input$wwtp_biobanksweden_arc_all
      d <- midasdashboard::biobanksweden_arc
      if(!is.null(input$wwtp_biobanksweden_arc_all)) {
        ds <- suppressMessages(amp_subset_samples(d, Plant %in% biobanksweden_arc_plants))
      } else
        ds <- d
    } else if(input$dataset == "CP332 (Egå+Marselisborg)") {
      CP332 <<- input$wwtp_CP332
      d <- midasdashboard::egaamarselisborg
      if(!is.null(input$wwtp_CP332)) {
        ds <- suppressMessages(amp_subset_samples(d, Plant %in% CP332))
      } else
        ds <- d
    }
    
    d$metadata <- d$metadata %>% 
      mutate(
        "Alle_anlæg" = if(input$dataset == "MiDAS OnlineKontrol")
          "Alle i datasæt + prøve type"
        else
          "Alle i datasæt"
      )
    
    if(input$genusfunction != "all") {
      ds$tax <- filter(
        ds$tax,
        grepl(
          paste0(
            as.character(midasdashboard::mfg_functions$Genus[midasdashboard::mfg_functions[,input$genusfunction] %in% c("POS")]),
            "$",
            collapse = "|"
          ),
          Genus
        )
      )
      rownames(ds$tax) <- ds$tax$OTU
      ds$abund <- ds$abund[rownames(ds$tax),]
    }
    
    out <- list(
      "complete" = ampvis2:::amp_rename(d),
      "subset" = ds
    )
    return(out)
  })
  
  data_subset <- reactive({
    d <- data()[["subset"]]
    if(!is.null(input$filtergenera)) {
      d$tax <- filter(
        d$tax, 
        grepl(
          paste0(
            as.character(input$filtergenera),
            "$",
            collapse = "|"
          ),
          Genus
        )
      )
      rownames(d$tax) <- d$tax$OTU
      d$abund <- d$abund[rownames(d$tax),]
    }
    return(d)
  })
  
  output$filtergenera <- renderUI({
    if(input$dataset == "MiDAS OnlineKontrol")
      req(input$samplecontent_midasok)
    selectizeInput(
      inputId = "filtergenera",
      label = "Filtrér til specifikke bakterier",
      choices = sort(
        str_remove_all(
          unique(
            as.character(data()[["subset"]][["tax"]][["Genus"]])
          ), 
          "^[kpcofgs]_*"
        )
      ),
      multiple = TRUE,
      options = list(placeholder = "Alle")
    )
  })
  
  ######################### LOGIN #########################
  auth <- reactiveValues(status = FALSE)
  observeEvent(input$login_button, {
    if((input$user_name == "MiDAS2017" & input$password == "spildevand") |
       (input$user_name == "aarhusvand" & input$password == "egaamarselisborg")) {
      auth$status <- TRUE
    } else {
      auth$status <- FALSE
    }
  })
  
  output$logintext <- renderUI({
    if(isTRUE(auth$status)) {
      shiny::tagList(
        wellPanel(
          h3("Velkommen til MiDAS dashboard"),
          tags$hr(),
          p("Dette er en online ", a(href = "https://shiny.rstudio.com/", "Shiny app"), " lavet til at give et indblik i mikrobiologien i danske renseanlæg siden år 2006. "),
          p("Al data i denne app er baseret på prøver fra danske renseanlæg, der er indsamlet og behandlet af Aalborg Universitet som en del af ", a(href="http://midasfieldguide.org", target = "_blank", "MiDAS"), " (Microbial Database of Activated Sludge) projektet, som har til formål at samle viden om de mikroorganismer, der spiller en central rolle i de biologiske processer i spildevandsrensning."),
          p("Start med at vælg et datasæt og ét eller flere anlæg til venstre. For en detaljeret guide til appen og dens funktioner se", downloadLink("guide_pdf", "denne PDF"), "."),
          br(),
          a(href="http://midasfieldguide.org", target = "_blank", img(src = "www/logobig.png"))
        )
      )
    }
  })
  
  ######################### SIDEBAR #########################
  output$sidebarUI <- renderMenu({
    sidebarMenu(
      menuItem(
        "Login",
        tabName = "login",
        icon = icon("user")
      ),
      if(isTRUE(auth$status)) {
        list(
          menuItem(
            "Anlæg info",
            tabName = "plantinfo",
            icon = icon("info")
          ),
          menuItem(
            text = "Bakteriesammensætning",
            tabName = "community",
            icon = icon("bug")
          ),
          menuItem(
            text = "Tidsserier",
            tabName = "timeseries",
            icon = icon("chart-line")
          ),
          menuItem(
            text = "Ordinering",
            tabName = "ordination",
            icon = icon("sitemap")
          ),
          tags$hr(),
          selectInput(
            inputId = "dataset",
            label = "Datasæt",
            choices = if(input$user_name == "aarhusvand") {
              c("MiDAS kvartalsprøver AS",
                "MiDAD kvartalsprøver AD (bakterier)",
                "MiDAD kvartalsprøver AD (archaea)",
                "BioBank AD (bakterier)", 
                "BioBank AD (archaea)", 
                "BioBank AS",
                "Sverige (bakterier)",
                "Sverige (archaea)",
                "CP332 (Egå+Marselisborg)"
              )
            } else {
              c(
                "MiDAS OnlineKontrol",
                "MiDAS kvartalsprøver AS",
                "MiDAD kvartalsprøver AD (bakterier)",
                "MiDAD kvartalsprøver AD (archaea)",
                "BioBank AD (bakterier)", 
                "BioBank AD (archaea)", 
                "BioBank AS",
                "Sverige (bakterier)",
                "Sverige (archaea)"
              )
            },
            selected = "Aktivt slam",
            multiple = FALSE
          ),
          # I don't know why there are more than one of these...
          conditionalPanel(
            condition = "input.dataset == 'MiDAS kvartalsprøver AS'",
            selectizeInput(
              inputId = "wwtp_midas",
              label = "Renseanlæg (blank for alle)",
              choices = sort(as.character(unique(midasdashboard::MiDAS$metadata$Plant))),
              selected = c("Aalborg West", "Aalborg East"),
              multiple = TRUE,
              options = list(placeholder = "Alle")
            )
          ),
          conditionalPanel(
            condition = "input.dataset == 'MiDAS OnlineKontrol'",
            tagList(
              selectizeInput(
                inputId = "samplecontent_midasok",
                label = "Prøve type",
                choices = sort(as.character(unique(midasdashboard::midasok$metadata$SampleContent))),
                multiple = TRUE,
                options = list(placeholder = "Alle")
              ),
              selectizeInput(
                inputId = "wwtp_midasok",
                label = "Renseanlæg (blank for alle)",
                choices = sort(as.character(unique(midasdashboard::midasok$metadata$Plant))),
                selected = c("Aalborg West", "Aalborg East"),
                multiple = TRUE,
                options = list(placeholder = "Alle")
              )
            )
          ),
          conditionalPanel(
            condition = "input.dataset == 'MiDAD kvartalsprøver AD (bakterier)'",
            selectizeInput(
              inputId = "wwtp_midad_bac",
              label = "Renseanlæg (blank for alle)",
              choices = sort(as.character(unique(midasdashboard::midad_bac$metadata$Plant))),
              selected = c("Aalborg West", "Aalborg East"),
              multiple = TRUE,
              options = list(placeholder = "Alle")
            )
          ),
          conditionalPanel(
            condition = "input.dataset == 'MiDAD kvartalsprøver AD (archaea)'",
            selectizeInput(
              inputId = "wwtp_midad_arc",
              label = "Renseanlæg (blank for alle)",
              choices = sort(as.character(unique(midasdashboard::midad_arc$metadata$Plant))),
              selected = c("Aalborg West", "Aalborg East"),
              multiple = TRUE,
              options = list(placeholder = "Alle")
            )
          ),
          conditionalPanel(
            condition = "input.dataset == 'BioBank AD (bakterier)'",
            selectizeInput(
              inputId = "wwtp_biobank_ad_bac",
              label = "Renseanlæg (blank for alle)",
              choices = sort(as.character(unique(midasdashboard::biobank_ad_bac$metadata$Plant))),
              selected = c("Aalborg West", "Aalborg East"),
              multiple = TRUE,
              options = list(placeholder = "Alle")
            )
          ),
          conditionalPanel(
            condition = "input.dataset == 'BioBank AD (archaea)'",
            selectizeInput(
              inputId = "wwtp_biobank_ad_arc",
              label = "Renseanlæg (blank for alle)",
              choices = sort(as.character(unique(midasdashboard::biobank_ad_arc$metadata$Plant))),
              selected = c("Aalborg_West", "Aalborg_East"),
              multiple = TRUE,
              options = list(placeholder = "Alle")
            )
          ),
          conditionalPanel(
            condition = "input.dataset == 'BioBank AS'",
            selectizeInput(
              inputId = "wwtp_biobank_all",
              label = "Renseanlæg (blank for alle)",
              choices = sort(as.character(unique(midasdashboard::biobank_as$metadata$Plant))),
              selected = c("Esbjerg West", "Esbjerg East"),
              multiple = TRUE,
              options = list(placeholder = "Alle")
            )
          ),
          conditionalPanel(
            condition = "input.dataset == 'Sverige (bakterier)'",
            selectizeInput(
              inputId = "wwtp_biobanksweden_bac_all",
              label = "Renseanlæg (blank for alle)",
              choices = sort(as.character(unique(midasdashboard::biobanksweden_bac$metadata$Plant))),
              selected = NULL,
              multiple = TRUE,
              options = list(placeholder = "Alle")
            )
          ),
          conditionalPanel(
            condition = "input.dataset == 'Sverige (archaea)'",
            selectizeInput(
              inputId = "wwtp_biobanksweden_arc_all",
              label = "Renseanlæg (blank for alle)",
              choices = sort(as.character(unique(midasdashboard::biobanksweden_arc$metadata$Plant))),
              selected = NULL,
              multiple = TRUE,
              options = list(placeholder = "Alle")
            )
          ),
          conditionalPanel(
            condition = "input.dataset == 'CP332 (Egå+Marselisborg)'",
            selectizeInput(
              inputId = "wwtp_CP332",
              label = "Renseanlæg (blank for alle)",
              choices = sort(as.character(unique(midasdashboard::egaamarselisborg$metadata$Plant))),
              selected = c("Aalborg West", "Aalborg East"),
              multiple = TRUE,
              options = list(placeholder = "Alle")
            )
          ),
          selectInput(inputId = "genusfunction",
                      label = "Filtrér til kendt Genus funktion",
                      choices = c("Alle bakterier" = "all",
                                  "Trådformende" = "FIL",
                                  "Ammonium oxiderende bakterier (AOB)" = "AOB",
                                  "Nitrit oxiderende bakterier (NOB)" = "NOB",
                                  "Polyfosfat akkumulerende bakterier (PAO)" = "PAO",
                                  "Glykogen akkumulerende bakterier (GAO)" = "GAO",
                                  "Anammox" = "Anammox"
                      ),
                      selected = "all",
                      multiple = FALSE
          ),
          #limit the height of the heatmap_filtergenera selectinput below:
          tags$style("#filtergenera ~ .selectize-control .selectize-input {max-height: 100px;overflow-y: auto;}"),
          uiOutput("filtergenera"),
          tags$hr(),
          div(
            style="text-align:center;position: absolute; bottom:2%; left: 40%;",
            tags$a(
              p(
                icon("github"),
                tags$br(),
                paste(
                  "Version:",
                  packageVersion("midasdashboard")
                  )
                ),
              href = "https://github.com/cmc-aau/midasdashboard",
              target="_blank"
            )
          ),
          conditionalPanel(
            condition="$('html').hasClass('shiny-busy')", 
            div(style="text-align:center",
                icon("sync"), 
                "Loading...")
          )        
        )
      }
    )
  })
  
  ######################### PLANTINFO #########################
  output$plantinfo <- DT::renderDataTable(
    data.table::fread(
      system.file(
        "plantinfo.csv",
        package = "midasdashboard",
        mustWork = TRUE
      )
    ) %>% 
      arrange(Anlæg),
    rownames = FALSE,
    options = list(
      pageLength = 50
    )
  )
  ######################### HEATMAP #########################
  output$heatmap_sort <- renderUI(
    selectizeInput(
      inputId = "heatmap_sort",
      label = "Sortér",
      choices = c(
        "Gennemsnit af valgte",
        unique(data_subset()$metadata[,input$heatmap_group_by])
      ),
      multiple = FALSE,
      width = "200px"
    )
  )
  
  heatmap <- eventReactive(input$render_heatmap, {
    heatmap <- NULL
    allheatmap <- NULL
    functions <- NULL
    #heatmap
    facet_by <- if(!input$heatmap_facet_by == "none") {input$heatmap_facet_by}
    if(input$heatmap_sort == "Gennemsnit af valgte") {
      order_by <- NULL
    } else {
      order_by <- input$heatmap_sort
    }
    
    heatmap <- amp_heatmap(data = data_subset(),
                           group_by = input$heatmap_group_by, 
                           facet_by = facet_by,
                           tax_aggregate = "Genus",
                           order_y_by = order_by,
                           tax_show = input$heatmap_tax_show,
                           plot_values = input$heatmap_plot_values,
                           plot_colorscale = "log10",
                           normalise = FALSE,
                           function_data = midasdashboard::mfg_functions,
                           functions = colnames(midasdashboard::mfg_functions[-1]),
                           color_vector = unlist(
                             strsplit(input$heatmap_colorvector, ",")
                           )
    )
    heatmap <- heatmap + 
      theme(legend.position = "none",
            axis.text.x = element_text(size = 14),
            axis.text.y = element_text(size = 12),
            axis.line = element_blank(),
            strip.text = element_text(angle = ifelse(isTRUE(input$heatmap_turnfacettext), 90, 0))
      )
    
    #allheatmap
    if(isTRUE(input$heatmap_showmean)) {
      d <- data()[["complete"]]
      pattern <- paste0(unique(heatmap$data$Display), "$", collapse = "|")
      pattern <- gsub("\\(", "\\\\(", pattern)
      pattern <- gsub("\\)", "\\\\)", pattern)
      d$tax <- filter(d$tax, grepl(pattern, Genus))
      rownames(d$tax) <- d$tax$OTU
      d$abund <- d$abund[rownames(d$tax),]
      allheatmap <- amp_heatmap(
        data = d,
        group_by = "Alle_anlæg",
        tax_aggregate = "Genus",
        order_y_by = levels(droplevels(heatmap$data$Display)),
        tax_show = input$heatmap_tax_show,
        plot_values = input$heatmap_plot_values,
        plot_colorscale = "log10",
        normalise = FALSE,
        function_data = midasdashboard::mfg_functions,
        color_vector = unlist(strsplit(input$heatmap_colorvector, ","))
      ) + 
        theme(legend.position = "none",
              axis.text.y = element_blank(), 
              axis.ticks.y = element_blank(),
              axis.text.x = element_text(size = 14),
              axis.line = element_blank()
        )
    }
    
    #functions
    if(isTRUE(input$heatmap_function)) {
      function_data <- midasdashboard::mfg_functions
      plotted_functions = colnames(function_data)[-1]
      # Retrieve the genus names from the plot
      names <- data.frame(do.call('rbind', strsplit(levels(droplevels(heatmap$data$Display)),'; ',fixed=TRUE)))
      names <- data.frame(Genus = names[,1])
      names$Genus <- as.character(names$Genus)
      
      # Define some nice colors for plotting
      POS <- "#31a354"
      VAR <- "orange"
      NEG <- "#f03b20"
      NT <- "grey90"
      
      # Make selection if not plotting all functions
      function_data <- function_data[,c("Genus",plotted_functions)]
      
      # Merge the genus and function information
      nameFunc <- merge(x = names, y = function_data, all.x = TRUE, all.y = FALSE) 
      nameFunc[is.na(nameFunc)] <- "NT"
      nameFuncM <- data.table::melt(
        data.table::data.table(nameFunc),
        id.vars = "Genus",
        value.name = "Value",
        variable.name = "Function"
      )
      nameFuncM$Value <- factor(nameFuncM$Value, levels = c("POS", "VAR", "NEG", "NT"))
      nameFuncM$Genus <- factor(nameFuncM$Genus, levels = names$Genus)
      
      # Generate the plot
      functions <- ggplot(nameFuncM, aes(x = Function, y = Genus, color = Value)) +
        geom_point(size = 4) +
        scale_color_manual(values = c(POS, VAR, NEG, NT), labels = c("Positiv", "Variabel", "Negativ", "Ikke testet"), drop = FALSE) +
        theme(axis.text.x = element_text(size = 12, color = "black", angle = 90, hjust = 1, vjust = 0.4),
              axis.text.y = element_blank(),
              axis.title = element_blank(),
              legend.title = element_blank(), 
              axis.ticks.length = unit(1, "mm"),
              axis.ticks = element_blank(),
              panel.background = element_blank(),
              panel.grid.major = element_line(color = "grey95"),
              legend.key = element_blank(),
              axis.line = element_blank()
        )
    }
    
    if(!isTRUE(input$heatmap_showmean) & !isTRUE(input$heatmap_function)) {
      plot <- heatmap
    } else if(isTRUE(input$heatmap_showmean) & !isTRUE(input$heatmap_function)) {
      plot <- cowplot::plot_grid(
        heatmap,
        allheatmap,
        ncol = 2,
        rel_widths = c(0.90, 0.10),
        align = "h",
        axis = "tb"
      )
    } else if(isTRUE(!input$heatmap_showmean) & isTRUE(input$heatmap_function)) {
      plot <- cowplot::plot_grid(
        heatmap,
        functions,
        ncol = 2,
        rel_widths = c(0.70, 0.30),
        align = "h",
        axis = "tb"
      )
    } else if(isTRUE(input$heatmap_showmean) & isTRUE(input$heatmap_function)) {
      plot <- cowplot::plot_grid(
        heatmap,
        allheatmap,
        functions,
        ncol = 3,
        rel_widths = c(0.65, 0.10, 0.25),
        align = "h",
        axis = "tb"
      )
    }
    
    #generate textmap
    textmap <- list()
    textmap$heatmap <- heatmap[["data"]][,c("Display", "Abundance", ".Group"), drop = FALSE] %>% 
      unique() %>%
      spread(key = ".Group", value = "Abundance") %>% 
      arrange(desc(droplevels(Display))) 
    colnames(textmap$heatmap)[1] <- "Genus"
    
    if(!is.null(allheatmap)) {
      textmap$allheatmap <- allheatmap[["data"]][,c("Display", "Abundance", ".Group"), drop = FALSE] %>% 
        unique() %>%
        spread(key = ".Group", value = "Abundance") %>% 
        arrange(desc(droplevels(Display))) %>% 
        select(-Display)
    }
    
    if(!is.null(functions)) {
      textmap$functions <- functions[["data"]] %>% 
        unique() %>% 
        spread(key = "Function", value = "Value") %>%
        arrange(desc(droplevels(Genus))) %>% 
        select(-Genus)
    }
    
    #return
    outlist <- list(heatmap = plot,
                    textmap = bind_cols(textmap))
    return(outlist)
  })
  
  output$heatmap <- renderPlot({
    heatmap()[["heatmap"]]
  })
  
  ######################### TIMESERIES #########################
  timeseries <- eventReactive(input$render_timeseries, {
    plot <- suppressWarnings(
      amp_timeseries(
        data = data_subset(),
        time_variable = "Date",
        tax_aggregate = "Genus",
        group_by = "Plant",
        normalise = FALSE,
        tax_show = input$timeseries_tax_show,
        split = TRUE,
        scales = if(isTRUE(input$timeseries_freey)) "free_y" else "fixed"
      )
    )
    plot <- plot + 
      scale_x_date(date_labels = "%Y", 
                   date_break = "1 year")
    plotlyplot <- ggplotly(plot) %>% 
      layout(margin = list(l = 100,
                           r = 50,
                           b = 100,
                           t = 20,
                           pad = 4
      )
      )
    plotlyplot$elementId <- NULL #To suppress warning spam
    return(plotlyplot)
  })
  
  output$timeseries <- renderPlotly({
    timeseries()
  })
  
  output$timeseries_svi <- renderPlotly({
    if(input$dataset == "Aktivt slam") {
      plot <- ggplot(data_subset()$metadata, aes(x=Date, y=SVI)) +
        geom_line(aes(group = Plant, color = Plant)) +
        geom_point(aes(group = Plant, color = Plant)) +
        scale_x_date(date_labels = "%Y", 
                     date_break = "1 year") +
        xlab("") +
        ylab("Slam Volumen Index") +
        theme_classic() +
        theme(axis.text.x = element_text(size = 10, vjust = 0.3, angle = 90),
              panel.grid.major.x = element_line(color = "grey90"),
              panel.grid.major.y = element_line(color = "grey90"),
              legend.title=element_blank()
        )
      
      plotlyplot <- ggplotly(plot) %>% 
        layout(margin = list(l = 70,
                             r = 50,
                             b = 50,
                             t = 20,
                             pad = 4
        )
        )
      plotlyplot$elementId <- NULL #To suppress warning spam
      return(plotlyplot)
    }
  })
  
  output$timeseries_dsvi <- renderPlotly({
    if(any(input$dataset %in% c("Aktivt slam", "BioBANK"))) {
      plot <- ggplot(data_subset()$metadata, aes(x=Date, y=DSVI)) +
        geom_line(aes(group = Plant, color = Plant)) +
        geom_point(aes(group = Plant, color = Plant)) +
        scale_x_date(date_labels = "%Y", 
                     date_break = "1 year") +
        xlab("") +
        ylab("Fortyndet Slam Volumen Index") +
        theme_classic() +
        theme(axis.text.x = element_text(size = 10, vjust = 0.3, angle = 90),
              panel.grid.major.x = element_line(color = "grey90"),
              panel.grid.major.y = element_line(color = "grey90"),
              legend.title=element_blank()
        )
      
      plotlyplot <- ggplotly(plot) %>% 
        layout(margin = list(l = 70,
                             r = 50,
                             b = 50,
                             t = 20,
                             pad = 4
        )
        )
      plotlyplot$elementId <- NULL #To suppress warning spam
      return(plotlyplot)
    }
  })
  
  ######################### ORDINATION #########################
  ordination <- eventReactive(input$render_ordination, {
    plot <- amp_ordinate(
      data_subset(),
      type = "PCA",
      transform = "hellinger",
      sample_color_by = "Plant",
      sample_colorframe = "Plant",
      sample_colorframe_label = "Plant",
      sample_trajectory = if(isTRUE(input$ord_trajectory)) "Date" else NULL,
      sample_trajectory_group = "Plant",
      sample_plotly = c("Plant", "Date")
    )
    return(plot)
  })
  
  output$ordination <- renderPlotly({
    plotly <- ggplotly(ordination()) %>%
      layout(showlegend = FALSE)
    plotly$elementId <- NULL #To suppress warning spam
    return(plotly)
  })
  
  ######################### DOWNLOAD HANDLERS #########################
  output$guide_pdf <- downloadHandler(
    filename = "MiDAS dashboard guide.PDF",
    content = function(file) {
      file.copy(system.file("app_guide.pdf", package = "midasdashboard"), file)
    }
  )
  
  output$RColorSheet <- downloadHandler(
    filename = "rcolors.pdf",
    content = function(file) {
      file.copy(system.file("rcolors.pdf", package = "midasdashboard"), file)
    }
  )
  
  output$saveTextmap <- downloadHandler(
    filename = "exported_heatmap.xlsx",
    content = function(file) {
      openxlsx::write.xlsx(heatmap()[["textmap"]], file)
    }
  )
}