From 7f69b26a496d5c0b25167c9305b9912758a6f400 Mon Sep 17 00:00:00 2001
From: Elias Projahn <elias@johrpan.de>
Date: Mon, 30 May 2022 21:59:40 +0200
Subject: [PATCH] Add outputs for custom genes

---
 R/plots.R  | 29 ++++++++++++++++++++++++++++-
 R/server.R | 49 ++++++++++++++++++++++++++++++++++++++++++++++++-
 R/ui.R     | 28 +++++++++++++++++++---------
 3 files changed, 95 insertions(+), 11 deletions(-)

diff --git a/R/plots.R b/R/plots.R
index 897ccd9..a253628 100644
--- a/R/plots.R
+++ b/R/plots.R
@@ -24,7 +24,7 @@ overview_plot <- function(ranked_data,
       yaxis = list(title = "Score")
     )
 
-  if (!is.null(highlighted_genes)) {
+  if (length(highlighted_genes) > 0) {
     figure <- figure |>
       plotly::add_markers(
         data = ranked_data[gene %chin% highlighted_genes],
@@ -44,6 +44,33 @@ overview_plot <- function(ranked_data,
   figure
 }
 
+#' Create a plot comparing some genes with the overall ranking.
+#'
+#' @param ranked_data The ranking to visualize.
+#' @param highlighted_genes Genes that will be compared.
+#'
+#' @return A `plotly` figure.
+#' @noRd
+box_plot <- function(ranked_data, highlighted_genes) {
+  data <- data.table::copy(ranked_data)
+  data[, group := data.table::fifelse(
+    gene %chin% highlighted_genes,
+    "Your genes",
+    "Other genes"
+  )]
+
+  plotly::plot_ly() |>
+    plotly::add_boxplot(
+      data = data,
+      x = ~score,
+      y = ~group,
+      boxpoints = FALSE
+    ) |> plotly::layout(
+      xaxis = list(title = "Score"),
+      yaxis = list(title = "")
+    )
+}
+
 #' Create plot showing the distribution of scores using `plotly`.
 #'
 #' @param ranked_data Data on genes with precomputed ranks.
diff --git a/R/server.R b/R/server.R
index 2ca7e05..90f9be1 100644
--- a/R/server.R
+++ b/R/server.R
@@ -1,6 +1,6 @@
 #' Server implementing the main user interface.
 #' @noRd
-server <- function(input, output) {
+server <- function(input, output, session) {
   ranked_data <- reactive({
     total_weight <- abs(input$cross_sample_weight) + abs(input$sd_expression)
     data <- data.table::copy(ubigen::genes)
@@ -10,6 +10,9 @@ server <- function(input, output) {
         input$sd_expression * sd_expression_normalized) /
         total_weight]
 
+    # Normalize scores to be between 0.0 and 1.0.
+    data[, score := (score - min(score)) / (max(score) - min(score))]
+
     data.table::setorder(data, -score)
     data[, rank := .I]
     data[, percentile := 1 - rank / max(rank)]
@@ -24,6 +27,50 @@ server <- function(input, output) {
     highlighted_genes = custom_genes()
   ))
 
+  observeEvent(custom_genes(),
+    { # nolint
+      if (length(custom_genes()) > 0) {
+        updateTabsetPanel(session, "custom_genes_panel", selected = "show")
+      } else {
+        updateTabsetPanel(session, "custom_genes_panel", selected = "hide")
+      }
+    },
+    ignoreNULL = FALSE
+  )
+
+  output$custom_genes_synopsis <- renderText({
+    comparison_gene_ids <- custom_genes()
+
+    if (length(comparison_gene_ids) > 1) {
+      reference <- ranked_data()[!gene %chin% comparison_gene_ids, score]
+      comparison <- ranked_data()[gene %chin% comparison_gene_ids, score]
+
+      p_value <- stats::wilcox.test(
+        x = comparison,
+        y = reference,
+        alternative = "greater"
+      )$p.value
+
+      reference_median <- stats::median(reference)
+      comparison_median <- stats::median(comparison)
+
+      HTML(glue::glue(
+        "The p-value for the alternative hypothesis that your genes have ",
+        "higher scores than other genes is <b>{format(round(p_value, ",
+        "digits = 4), nsmall = 4, scientific = FALSE)}</b>. This value was ",
+        "computed using a Wilcoxon rank sum test. The median score of your ",
+        "genes is <b>{format(round(comparison_median, digits = 2), ",
+        "nsmall = 2, scientific = FALSE)}</b> compared to a median score of ",
+        "<b>{format(round(reference_median, digits = 2), nsmall = 2, ",
+        "scientific = FALSE)}</b> of the other genes."
+      ))
+    }
+  })
+
+  output$custom_genes_boxplot <- plotly::renderPlotly(
+    box_plot(ranked_data(), custom_genes())
+  )
+
   output$scores_plot <- plotly::renderPlotly(scores_plot(
     ranked_data(),
     highlighted_genes = custom_genes()
diff --git a/R/ui.R b/R/ui.R
index 889ab44..f0d1147 100644
--- a/R/ui.R
+++ b/R/ui.R
@@ -20,12 +20,12 @@ ui <- function() {
           selectInput(
             "cross_sample_metric",
             verticalLayout(
-               strong("Expression across samples"),
-               paste0(
-                 "Proportion samples in which the gene is expressed above the ",
-                 "selected threshold. Select a method and a weight for the ",
-                 "final score."
-               )
+              strong("Expression across samples"),
+              paste0(
+                "Proportion samples in which the gene is expressed above the ",
+                "selected threshold. Select a method and a weight for the ",
+                "final score."
+              )
             ),
             list(
               "Above 95th percentile" = "above_95",
@@ -55,10 +55,20 @@ ui <- function() {
         ),
         mainPanel(
           width = 9,
-          h3("Distribution of scores"),
-          h4("Overview"),
+          h3("Overview"),
           plotly::plotlyOutput("overview_plot", height = "200px"),
-          h4("Focus on top genes"),
+          tabsetPanel(
+            id = "custom_genes_panel",
+            type = "hidden",
+            tabPanelBody("hide"),
+            tabPanelBody(
+              "show",
+              h3("Your genes"),
+              htmlOutput("custom_genes_synopsis"),
+              plotly::plotlyOutput("custom_genes_boxplot")
+            )
+          ),
+          h3("Focus on top genes"),
           div(paste0(
             "Click or drag within the figure to select genes of ",
             "interest."