app_normalization.R

#~~~~~~~~~~~~~~~~~~~ App aiding in normalization of GeoMx-data ~~~~~~~~~~~~~~~~~~~~
# Author: Elias Carlsson
# Email: eeliascarlsson@gmail.com
# Description: An app making protein boxplot, scatterplots, correlation matrices and PCA-plots
# Part of apps_for_geomx project
#~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

# Cleaning
rm(list = ls())
cat("\014")
graphics.off()

# Sourcing
library('shiny')
library("corrplot")
source("./functions/load_data.R")
source("./functions/normalization_functions.R")

# Loading data for app
load("./data/setup.RData")
allprot = proteins
rm(proteins)
location1 = loc[1]
data_sel = load_data(location = location1)

# UI
ui <- fluidPage(
  
  # Sidebar
  sidebarLayout(
    sidebarPanel(
      style = "position:fixed;width:inherit;height: 95vh; overflow-y: auto;",
      width = 3,
      h1("app_normalization"),
      h3("Choose dataset"),
      
      selectInput(inputId ="location_norm", label = strong("NormalyzedDE Data set"),
                  choices = loc_nDE,
                  selected = loc_nDE[2]), 
      
      uiOutput("location"),
      
      h3("Filter data"),
      
      selectInput(inputId ="filter", label = strong("Filter type"),
                  multiple = T,
                  choices = vec_type,
                  selected = ""), 
      
      uiOutput("filter_ann"),
      
      uiOutput("no_samples"),
      
      h3("Plot options"),
      
      selectInput(inputId ="type", label = strong("Color by"),
                  multiple = F,
                  choices = vec_type,
                  selected = "Main_class"),
      
      selectInput(inputId ="cvec", label = strong("Colors"),
                  multiple = T, 
                  choices = c("grey","blue","red","orange","purple","lightblue"),
                  selected = c("grey","blue","red")),
      
      uiOutput("color_one_type"),
      
    ),
    
    # Main panel
    mainPanel(
      tabsetPanel(id = "tabs",
        tabPanel(title = "Boxplot", value = "BP",
                 fluidRow(align = "center", h3("Protein boxplot")),
                 
                 fluidRow(align = "center", 
                          column(3, radioButtons(inputId = "snryn", label = strong("Type of plot:"), 
                                                 choices = c("Expression", "SNR"),
                                                 selected = "Expression",
                                                 inline = T,)),
                          column(3, sliderInput(inputId = "dotsize", "Dotsize",
                                                min = 0.5, max = 2.5, value = 1, step = 0.5
                          )),
                          column(3, selectInput(inputId = "proteins", label = strong("Select Proteins"),
                                                multiple = F,
                                                choices = c("All", "Control and HK", "Custom"),
                                                selected = "Control and HK")),
                          column(3, uiOutput("proteins_custom"))),
                 
                 fluidRow(align = "center", 
                          column(12, div(style = "height:500px", plotOutput("protein_boxplot"))))),
        
        tabPanel(title = "Scatterplots", value = "SC",
                 fluidRow(align = "center", h3("Scatterplots")), 
                 
                 fluidRow(align = "center", 
                          column(6, plotOutput("scatterplot1")),
                          column(6, plotOutput("scatterplot2"))),
                 fluidRow(align = "center",
                          column(12,plotOutput("neg_scatterplot")))),
        
        tabPanel(title = "PCA analysis", value = "PCA",
                 fluidRow(align = "center", h3("PCA-plot in three dimensions")),
                 fluidRow(align = "center",
                          column(12, selectInput(inputId = "biomarker", label = strong("Select biomarker to color by"), 
                                                 multiple = F,
                                                 choices = c(allprot),
                                                 selected = "GAPDH"))),
                 fluidRow(align = "center", plotOutput("PCA")),
                 
                 fluidRow(align = "center", h3("Scree-plot")),
                 fluidRow(align = "center",
                          sliderInput(inputId = "co", "No. PC",
                                            min = 1, max = 78 , value = 10, step = 1)),
                 fluidRow(align = "center",
                          column(12, plotOutput("scree"))),
                 ),
        tabPanel(title = "Correlation Matrix", value = "cmat",
                 fluidRow(align = "center", h3("Correlation matrix (p-values)")),
                 fluidRow(align = "center",
                          column(12, selectizeInput(inputId = "proteins_corrmat", label = strong("Select proteins"),
                                                 multiple = T,
                                                 choices = unique(allprot)),
                                                 selected = unique(allprot))),
                 fluidRow(align = "center",
                          column(12, plotOutput("cormat")))
                 )),
      
    )
  )
)

# Server
server <- function(input, output) {
  location_norm <- reactive({
    location_norm <- req(input$location_norm)
  })
  
  output$location <- renderUI({
    if (location_norm()==0){
      selectInput(inputId = "location", label = strong("Data set"),
                  choices = loc,
                  selected = loc[1])
    }
  })
  
  location <- reactive({
    if (location_norm()==0){
      location <- req(input$location)
    } else {
      location <- loc[1]
    }
  })
  
  filter <- reactive({
    filter = input$filter
  })
  
  output$filter_ann <- renderUI({
    if (!is.null(filter())){
      filter = filter()
      output <- tagList()
      for (i in 1:length(filter)){
        output[[i]] <- selectInput(inputId = filter[i], label = strong(filter[i]),
                                   multiple = T, 
                                   choices = sort(unique(data_sel[[filter[i]]])), 
                                   selected = NA)  
      }
      output
    }
  })
  
  filter_list <- reactive({
    l = list()
    filter = filter()
    for (i in 1:length(filter)){
      filter_ann = req(input[[filter[i]]])
      l[[i]] = filter_ann
    }
    filter_list = l
  })
  
  proteins <- reactive({
    proteins <- req(input$proteins)

  })
  
  output$proteins_custom <- renderUI({
    if (proteins() == "Custom"){
    selectInput(inputId = "proteins_custom", label = strong("Select custom proteins"),
                multiple = T,
                choices = allprot,
                selected = unique(cn))
    }
  })
  
  proteins_custom <- reactive ({
    if (proteins() == "All"){
      proteins_custom = allprot
    } else if (proteins() == "Control and HK"){
      proteins_custom = unname(all_of(cn))
    } else if (proteins() == "Custom"){
      proteins_custom = req(input$proteins_custom)
    }
  })
  
  type <- reactive({
    type = req(input$type)
  })
  
  cvec <- reactive({
    cvec = req(input$cvec)
  })
  
  output$color_one_type <- renderUI({
    selectInput(inputId = "color_one_type", label = strong("Color one annotation to all"),
                choices = c("None", data_sel[[type()]] %>% unique()))
  })
  
  color_one_type <- reactive({
    color_one_type <- req(input$color_one_type)
  })
  
  my_df <- reactive({
    df = load_data(location(), location_norm())
    if (!is.null(filter())){
      df = filter_data(df, filter(), filter_list())
    }
    
    type = type()
    if (!color_one_type() == "None"){
      for (i in 1:length(df[[type]])){
        if (!df[[type]][i] == color_one_type()){
          df[[type]][i] = "All other"
        }
      }
    }
    
    my_df = df

  })
  
  
  output$no_samples <- renderUI({
    helpText(paste(length(my_df()[[1]]), "samples selected"))
  })
  
  dotsize <- reactive({
    dotsize = req(input$dotsize)
  })
  
  
  snryn <- reactive({
    snryn = req(input$snryn)
  })
  
  tabstatus <- reactive({
    tabstatus <- req(input$tabs)
  })
  
  output$protein_boxplot <- renderPlot({
    if (tabstatus() == "BP"){
    if (snryn() == "Expression"){
    protein_boxplot(my_df = my_df(), 
                    type = type(), 
                    proteins = proteins_custom(),
                    cvec = cvec(),
                    dotsize = dotsize())
    } else {
      protein_boxplot_snr(my_df = my_df(), 
                      type = type(), 
                      proteins = proteins_custom(),
                      cvec = cvec(),
                      dotsize = dotsize())      
    }
  }}, height = 500, width = 1000 ) 
  
  output$scatterplot1 <- renderPlot({
    if (tabstatus() == "SC"){
    outlist = scatterplot_HK_Neg(my_df = my_df(),
                                 type = type(),
                                 cvec = cvec())
    outlist[[1]]
    }
  }, height = 400, width = 500)
  
  output$scatterplot2 <- renderPlot({
    if (tabstatus() == "SC"){
    outlist = scatterplot_HK_Neg(my_df = my_df(),
                                 type = type(),
                                 cvec = cvec())
    outlist[[2]]
  }}, height = 400, width = 500)
  
  output$neg_scatterplot <- renderPlot({
    if (tabstatus() == "SC"){
      scatterplot_Norm_neg(my_df = my_df())
      
    }
  }, height = 200, width = 1100)

  
  co <- reactive({
    co = req(input$co)
  })
  
  output$scree <- renderPlot({
    if (tabstatus() == "PCA"){
    scree_plot(my_df = my_df(),
               co = co())
  }}, height = 300, width = 800)
  
  biomarker <- reactive({
    biomarker = req(input$biomarker)
  })
  
  output$PCA <- renderPlot({
    if (tabstatus() == "PCA"){
        out = PCA_scatterplot_3d(my_df = my_df(),
                                 type = type(),
                                 biomarker = biomarker(),
                                 cvec = cvec()) 
        out[[1]]/out[[2]]
      }
  },height = 400, width = 500)
  
  proteins_corrmat <- reactive({
    proteins_corrmat <- req(input$proteins_corrmat)
  })
  
  output$cormat <- renderPlot({
    cormat(my_df(),proteins_corrmat())
  }, height = 500, width = 500)
}

# Run the application 
shinyApp(ui = ui, server = server)