From 363dee2cbdf4ab5083664902f6637e919f5acc79 Mon Sep 17 00:00:00 2001
From: Elias Projahn <elias@johrpan.de>
Date: Thu, 26 Aug 2021 12:50:03 +0200
Subject: [PATCH] Enhance scatter plot

---
 scatter_plot.R | 82 +++++++++++++++++++++++++++++++++++---------------
 server.R       |  2 +-
 2 files changed, 58 insertions(+), 26 deletions(-)

diff --git a/scatter_plot.R b/scatter_plot.R
index 1ed1e25..78d5f8f 100644
--- a/scatter_plot.R
+++ b/scatter_plot.R
@@ -2,31 +2,63 @@ library(data.table)
 library(ggplot2)
 
 #' Draw a scatter plot containing gene positions.
-scatter_plot <- function(gene_ids, data) {
-    plot <- ggplot() +
-        scale_x_discrete(
-            name = "Species",
-            breaks = data$species$id,
-            labels = data$species$label
-        ) +
-        scale_y_continuous(name = "Distance to telomeres [Mbp]")
-
-    colors <- rainbow(length(gene_ids))
-
-    for (i in seq_len(length(gene_ids))) {
-        gene_id <- gene_ids[i]
-
-        plot <- plot +
-            geom_point(
-                data$distances[gene == gene_id],
-                mapping = aes(
-                    x = species,
-                    y = distance / 1000000,
-                ),
-                color = colors[i],
-                size = 4
-            )
+#'
+#' @param input Input data from [`load_input()`].
+#' @param results Results from [`process_input()`].
+scatter_plot <- function(gene_ids, input, results) {
+    if (length(gene_ids) < 1) {
+        return(ggplot())
     }
 
-    plot
+    # Exclude species with naturally or artificially short chromosomes as well
+    # as non-replicatively aging species.
+    # TODO: Sync with process_input().
+    species <- input$species[
+        median_distance >= 7500000 & group == "replicative"
+    ]
+
+    species_ids <- species[, id]
+
+    data <- merge(
+        input$genes[id %in% gene_ids, .(id, name)],
+        input$distances[species %in% species_ids],
+        by.x = "id", by.y = "gene"
+    )
+
+    for (gene_id in gene_ids) {
+        cluster_species <- unlist(results[gene == gene_id, cluster_species])
+        data[id == gene_id, in_cluster := species %in% cluster_species]
+    }
+
+    ggplot(data) +
+        scale_x_discrete(
+            name = "Species",
+            breaks = species$id,
+            labels = species$label
+        ) +
+        scale_y_continuous(
+            name = "Distance to telomeres [Mbp]",
+            limits = function(x) {
+                if (x[2] < 15) {
+                    c(0, 15)
+                } else {
+                    x
+                }
+            }
+        ) +
+        scale_color_discrete(name = "Gene") +
+        scale_shape_discrete(
+            name = "Part of cluster",
+            breaks = c(TRUE, FALSE),
+            labels = c("Yes", "No")
+        ) +
+        geom_point(
+            mapping = aes(
+                x = species,
+                y = distance / 1000000,
+                color = name,
+                shape = in_cluster
+            ),
+            size = 5
+        )
 }
\ No newline at end of file
diff --git a/server.R b/server.R
index 8b8fddd..ea65a45 100644
--- a/server.R
+++ b/server.R
@@ -37,6 +37,6 @@ server <- function(input, output) {
 
     output$scatter <- renderPlot({
         gene_ids <- filtered()[input$genes_rows_selected, gene]
-        scatter_plot(gene_ids, data)
+        scatter_plot(gene_ids, data, results)
     })
 }
\ No newline at end of file