Merge pull request #84 from databio/staging

bug fixes

.gitignore

@@ -13,3 +13,4 @@ apps/LOLAweb/cache
 apps/LOLAweb/shinylog
 apps/LOLAweb/results
 apps/LOLAweb/userSets
+apps/LOLAweb/scratch.R

apps/LOLAweb/app.R

@@ -209,10 +209,8 @@ ui <- list(
                column(DT::dataTableOutput("res"), width = 12)
                ),
       tabPanel("Run summary",
-               conditionalPanel(condition = "output.res",
                                 h4("Run summary"),
                                 tableOutput("run_sum"), style = "font-size:18px;")
-               )
            ),
       id = "result-tabs")))
   )
@@ -590,15 +588,42 @@ server <- function(input, output, session) {
         # need to make sure user set is discrete even if coded as number
         resRedefined$userSet = as.character(resRedefined$userSet)
 
+        gd_data <- utils::data(package="GenomicDistributions")$results[,"Item"]
+
         # calculate distribution over chromosomes for plotting
-        genDist = aggregateOverGenomeBins(userSets, input$refgenome)
+        if (paste0("chromSizes_", input$refgenome) %in% gd_data) {
+
+          genDist = aggregateOverGenomeBins(userSets, input$refgenome)
+
+        } else {
+
+          genDist = NULL
+
+        }
 
         # calculate distances to TSSs
-        TSSDist = TSSDistance(userSets, input$refgenome)
+        if (paste0("TSS_", input$refgenome) %in% gd_data) {
+
+          TSSDist = TSSDistance(userSets, input$refgenome)
+
+
+        } else {
+
+          TSSDist = NULL
+
+        }
 
         # distribution of overlaps for a query set to genomic partitions
-        gp = genomicPartitions(userSets, input$refgenome)
+        if (paste0("geneModels_", input$refgenome) %in% gd_data) {
+
+          gp = genomicPartitions(userSets, input$refgenome)
 
+        } else {
+
+          gp = NULL
+
+        }
+
       })
 
       run_sum <- 
@@ -1140,7 +1165,17 @@ server <- function(input, output, session) {
 
     genDist <- rawdat_res$genDist
 
-    plotGenomeAggregate(genomeAggregate = genDist)
+    if (is.null(genDist)) {
+
+      NULL
+
+    } else {
+
+      plotGenomeAggregate(genomeAggregate = genDist) +
+        guides(fill=guide_legend(title="User set"),
+               col = guide_legend(title="User set"))
+
+    }
 
   }
 
@@ -1158,8 +1193,18 @@ server <- function(input, output, session) {
 
     TSSDist <- rawdat_res$TSSDist
 
-    plotFeatureDist(TSSDist, featureName="TSS")
-
+    if (is.null(TSSDist)) {
+
+      NULL 
+
+    } else {
+
+      plotFeatureDist(TSSDist, featureName="TSS") +
+        guides(fill=guide_legend(title="User set"),
+               col = guide_legend(title="User set"))
+
+    }
+
   }
 
   output$dist_plot <- renderPlot({

apps/LOLAweb/misc.R

@@ -34,10 +34,10 @@ plot_input <- function(res, metric, ylabel, sortcol) {
   if(grepl("rnk", sortcol, ignore.case = TRUE)) {
 
     # need to order data frame by sort col if it's a rank
-    dat <- res[order(as.data.frame(res)[,sortcol]), ]
+    res <- res[order(as.data.frame(res)[,sortcol]), ]
 
     # now construct base layer for plot with reverse on the sort
-    p <- ggplot(dat, aes(reorder(axis_label, 
+    p <- ggplot(res, aes(reorder(axis_label, 
       rev(eval(parse(text = sortcol)))), eval(parse(text = metric)), fill = userSet, group = id))
 
   } else {
@@ -48,14 +48,15 @@ plot_input <- function(res, metric, ylabel, sortcol) {
   }
 
   # Make label size dynamic based on number of entries in the table.
-  label_size = 9 + ( 8 * (1 - NROW(dat)/50) )
+  label_size = 9 + ( 8 * (1 - NROW(res)/50) )
   p +
     geom_bar(stat="identity", position="dodge") +
     xlab("Description") +
     ylab(ylabel) +
     coord_flip() +
     theme_ns() +
-    theme(axis.text=element_text(size=label_size))
+    theme(axis.text=element_text(size=label_size)) +
+    guides(fill=guide_legend(title="User set"))
 
 }