Add basic filters to the UI

server.R

@@ -9,16 +9,21 @@ source("util.R")
 
 data <- run_cached("input", load_input, "input")
 results <- run_cached("results", process_input, data)
+merged <- merge(results, data$genes, by.x = "gene", by.y = "id")
+setorder(merged, -cluster_length)
 
 server <- function(input, output) {
-    filtered <- results[cluster_length >= 10]
+    filtered <- reactive({
-    merged <- merge.data.table(filtered, data$genes, by.x = "gene", by.y = "id")
+        merged[
-    setorder(merged, -cluster_length)
+            cluster_length >= input$length &
+                cluster_mean >= input$range[1] * 1000000 &
+                cluster_mean <= input$range[2] * 1000000
+        ]
+    })
 
     output$genes <- renderDT({
         datatable(
-            merged[, .(.I, name, chromosome, cluster_length, cluster_mean)],
+            filtered()[, .(.I, name, chromosome, cluster_length, cluster_mean)],
-            rownames = FALSE,
             colnames = c(
                 "Rank",
                 "Gene",
@@ -31,7 +36,7 @@ server <- function(input, output) {
     })
 
     output$scatter <- renderPlot({
-        gene_ids <- merged[input$genes_rows_selected, gene]
+        gene_ids <- filtered()[input$genes_rows_selected, gene]
         scatter_plot(gene_ids, data)
     })
 }

ui.R

@@ -7,6 +7,21 @@ ui <- fluidPage(
         position = "right",
         sidebarPanel(
             h3("Candidate selection"),
+            sliderInput(
+                "range",
+                "Gene position (Mbp)",
+                min = 0,
+                max = 50,
+                value = c(0, 15),
+                step = 0.1
+            ),
+            sliderInput(
+                "length",
+                "Minimum cluster size",
+                min = 0,
+                max = 30,
+                value = 10
+            ),
             DTOutput("genes"),
             width = 3
         ),