WT: Plot updates

Author: williamthistle

R/map_cell_type_from_reference.R

@@ -406,12 +406,12 @@ MapCellTypes_RNA <- function(sc_obj, reference, reference_cell_type_attribute =
         sample_count <- length(unique(sc_obj$sample))
         cell_count <- length(sc_obj$cell_name)
         # Plot by majority vote cell type
-        current_title <- paste0("RNA Data Integration \n (By Majority Vote Cell Type) \n (", sample_count, " Samples, ", cell_count, " Cells)")
+        current_title <- paste0("RNA Data Integration\n(By Majority Vote Cell Type)\n(", sample_count, " Samples, ", cell_count, " Cells)")
         print_UMAP_RNA(sc_obj, file_name = "Final_RNA_UMAP_by_Majority_Vote_Cell_Type.png",
                        group_by_category = "predicted_celltype_majority_vote", output_dir = output_dir, title = current_title,
                        log_flag = log_flag)
         # Plot by raw predicted cell type
-        current_title <- paste0("RNA Data Integration \n (By Raw Predicted Cell Type) \n (", sample_count, " Samples, ", cell_count, " Cells)")
+        current_title <- paste0("RNA Data Integration\n(By Raw Predicted Cell Type)\n(", sample_count, " Samples, ", cell_count, " Cells)")
         print_UMAP_RNA(sc_obj, file_name = "Final_RNA_UMAP_by_Raw_Predicted_Cell_Type.png",
                        group_by_category = "predicted.id", output_dir = output_dir, title = current_title,
                        log_flag = log_flag)
@@ -503,15 +503,15 @@ MapCellTypes_ATAC <- function(proj, reference, reference_cell_type_attribute = "
         }
         pal <- paletteDiscrete(values = proj$predictedGroup)
         p1 <- plotEmbedding(ArchRProj = proj, colorBy = "cellColData", name = "predictedGroup", embedding = "UMAP", pal = pal, force = TRUE, keepAxis = TRUE) +
-          ggplot2::ggtitle(paste0("ATAC Data After Integration (By Raw Predicted Cell Type) \n ", sample_text)) + ggplot2::theme(plot.title = ggplot2::element_text(size=18))
+          ggplot2::ggtitle(paste0("ATAC Data Integration\n(By Raw Predicted Cell Type)\n", sample_text)) + ggplot2::theme(plot.title = ggplot2::element_text(size=18), legend.text = ggplot2::element_text(size=10))
         ggplot2::ggsave(filename = paste0(output_dir, "Final_ATAC_UMAP_by_Raw_Predicted_Cell_Type.png"), plot = p1, device = "png", width = 8, height = 8, units = "in")
         ArchR::plotPDF(p1, name = "Final_ATAC_UMAP_by_Raw_Predicted_Cell_Type", ArchRProj = proj, addDOC = FALSE, width = 5, height = 5)
         # We have to perform majority voting with a different cluster attribute if Harmony was not run
         # (due to only having one batch)
         proj <- MajorityVote_ATAC(proj, log_flag)
         pal <- paletteDiscrete(values = proj$Cell_type_voting)
         p2 <- ArchR::plotEmbedding(ArchRProj = proj, colorBy = "cellColData", name = "Cell_type_voting", embedding = "UMAP", pal = pal, force = TRUE, keepAxis = TRUE) +
-          ggplot2::ggtitle(paste0("ATAC Data After Integration (By Majority Vote Cell Type) \n ", sample_text)) + ggplot2::theme(plot.title = ggplot2::element_text(size=18))
+          ggplot2::ggtitle(paste0("ATAC Data Integration\n(By Majority Vote Cell Type)\n", sample_text)) + ggplot2::theme(plot.title = ggplot2::element_text(size=18), legend.text = ggplot2::element_text(size=10))
         ggplot2::ggsave(filename = paste0(output_dir, "Final_ATAC_UMAP_by_Majority_Vote_Cell_Type.png"), plot = p2, device = "png", width = 8, height = 8, units = "in")
         ArchR::plotPDF(p1,p2, name = "UMAP_Cell_Type_Label_Plots", ArchRProj = proj, addDOC = FALSE, width = 5, height = 5)
       }

R/process_batches.R

@@ -367,13 +367,13 @@ IntegrateByBatch_ATAC <- function(proj, output_dir = getwd(), exit_with_code = F
       }
       # Plot integrated UMAPs
       p1 <- ArchR::plotEmbedding(ArchRProj = proj, colorBy = "cellColData", name = "seurat_clusters", embedding = "UMAP", force = TRUE, keepAxis = TRUE) +
-        ggplot2::ggtitle(paste0("ATAC Data After Integration (By Clusters) \n ", sample_text)) + ggplot2::theme(plot.title = ggplot2::element_text(size=18))
+        ggplot2::ggtitle(paste0("ATAC Data Integration\n(By Clusters)\n", sample_text)) + ggplot2::theme(plot.title = ggplot2::element_text(size=18), legend.text = ggplot2::element_text(size=10))
       ggplot2::ggsave(filename = paste0(output_dir, "Final_ATAC_UMAP_by_Clusters.png"), plot = p1, device = "png", width = 8, height = 8, units = "in")
       p2 <- ArchR::plotEmbedding(ArchRProj = proj, colorBy = "cellColData", name = "Sample", embedding = "UMAP", force = TRUE, keepAxis = TRUE) +
-        ggplot2::ggtitle(paste0("ATAC Data After Integration (By Sample) \n ", sample_text)) + ggplot2::theme(plot.title = ggplot2::element_text(size=18))
+        ggplot2::ggtitle(paste0("ATAC Data Integration\n(By Sample)\n", sample_text)) + ggplot2::theme(plot.title = ggplot2::element_text(size=18), legend.text = ggplot2::element_text(size=10))
       ggplot2::ggsave(filename = paste0(output_dir, "Final_ATAC_UMAP_by_Sample.png"), plot = p2, device = "png", width = 8, height = 8, units = "in")
       p3 <- ArchR::plotEmbedding(ArchRProj = proj, colorBy = "cellColData", name = "TSSEnrichment", embedding = "UMAP", force = TRUE, keepAxis = TRUE) +
-        ggplot2::ggtitle(paste0("ATAC Data After Integration (By TSS Enrichment) \n ", sample_text)) + ggplot2::theme(plot.title = ggplot2::element_text(size=18))
+        ggplot2::ggtitle(paste0("ATAC Data Integration\n(By TSS Enrichment)\n", sample_text)) + ggplot2::theme(plot.title = ggplot2::element_text(size=18), legend.key.size = ggplot2::unit(1, "cm"), legend.text = ggplot2::element_text(size=10))
       ggplot2::ggsave(filename = paste0(output_dir, "Final_ATAC_UMAP_by_TSSEnrichment.png"), plot = p3, device = "png", width = 8, height = 8, units = "in")
       ArchR::plotPDF(p1,p2,p3, name = "UMAP_Final_Integrated_Plots", ArchRProj = proj, addDOC = FALSE, width = 5, height = 5)
       return(proj)
@@ -426,12 +426,12 @@ VisualizeIntegration <- function(sc_obj, output_dir = getwd(), exit_with_code =
       sample_count <- length(unique(sc_obj$sample))
       cell_count <- length(sc_obj$cell_name)
       # Plot by cluster
-      current_title <- paste0("RNA Data Integration (By Cluster) \n (", sample_count, " Samples, ", cell_count, " Cells)")
+      current_title <- paste0("RNA Data Integration\n(By Cluster)\n(", sample_count, " Samples, ", cell_count, " Cells)")
       print_UMAP_RNA(sc_obj, file_name = "Final_RNA_UMAP_by_Cluster.png",
                      group_by_category = "seurat_clusters", output_dir = output_dir, title = current_title,
                      log_flag = log_flag)
       # Plot by sample
-      current_title <- paste0("RNA Data Integration (By Sample) \n (", sample_count, " Samples, ", cell_count, " Cells)")
+      current_title <- paste0("RNA Data Integration\n(By Sample)\n(", sample_count, " Samples, ", cell_count, " Cells)")
       print_UMAP_RNA(sc_obj, file_name = "Final_RNA_UMAP_by_Sample.png",
                      group_by_category = "sample", output_dir = output_dir, title = current_title,
                      log_flag = log_flag)

R/process_multiome.R

@@ -33,13 +33,13 @@ FindMultiomeOverlap <- function(sc_obj, proj, data_modality = "RNA", output_dir
           sample_text <- paste0("(", num_samples, " Samples, ", num_cells, " Cells)")
         }
         p1 <- ArchR::plotEmbedding(ArchRProj = proj, colorBy = "cellColData", name = "Sample", embedding = "UMAP", force = TRUE, keepAxis = TRUE) +
-          ggplot2::ggtitle(paste0("Multiome Overlap for ATAC Data (By Sample) \n ", sample_text)) + ggplot2::theme(plot.title = ggplot2::element_text(size=18))
+          ggplot2::ggtitle(paste0("Multiome Overlap for ATAC Data\n(By Sample)\n", sample_text)) + ggplot2::theme(plot.title = ggplot2::element_text(size=18), legend.text = ggplot2::element_text(size=10))
         ggplot2::ggsave(filename = paste0(output_dir, "Final_Multiome_Overlap_ATAC_UMAP_by_Sample.png"), plot = p1, device = "png", width = 8, height = 8, units = "in")
       } else if(data_modality == "RNA") {
         sc_obj <- sc_obj[,ArchR::getCellNames(proj)]
         sample_count <- length(unique(sc_obj$sample))
         cell_count <- length(sc_obj$cell_name)
-        current_title <- paste0("Multiome Overlap for RNA Data (By Sample) \n (", sample_count, " Samples, ", cell_count, " Cells)")
+        current_title <- paste0("Multiome Overlap for RNA Data\n(By Sample)\n(", sample_count, " Samples, ", cell_count, " Cells)")
         print_UMAP_RNA(sc_obj, file_name = "Final_Multiome_Overlap_RNA_UMAP_by_Sample.png",
                        group_by_category = "sample", output_dir = output_dir, title = current_title,
                        log_flag = log_flag)
@@ -98,16 +98,16 @@ TransferRNALabels <- function(sc_obj, proj, RNA_output_dir = getwd(), ATAC_outpu
         sample_text <- paste0("(", num_samples, " Samples, ", num_cells, " Cells)")
       }
       p1 <- ArchR::plotEmbedding(ArchRProj = proj, colorBy = "cellColData", name = "Cell_type_voting_RNA", embedding = "UMAP", force = TRUE, keepAxis = TRUE) +
-        ggplot2::ggtitle(paste0("Multiome Overlap for ATAC Data \n (By RNA Majority Vote Cell Type) \n ", sample_text)) + ggplot2::theme(plot.title = ggplot2::element_text(size=18))
+        ggplot2::ggtitle(paste0("Multiome Overlap for ATAC Data\n(By RNA Majority Vote Cell Type)\n", sample_text)) + ggplot2::theme(plot.title = ggplot2::element_text(size=18), legend.text = ggplot2::element_text(size=10))
       ggplot2::ggsave(filename = paste0(ATAC_output_dir, "Final_Multiome_Overlap_ATAC_UMAP_by_RNA_Majority_Vote_Cell_Type_Labels.png"), plot = p1, device = "png", width = 8, height = 8, units = "in")
       p2 <- ArchR::plotEmbedding(ArchRProj = proj, colorBy = "cellColData", name = "Cell_type_voting", embedding = "UMAP", force = TRUE, keepAxis = TRUE) +
-        ggplot2::ggtitle(paste0("Multiome Overlap for ATAC Data \n (By ATAC Majority Vote Cell Type) \n ", sample_text)) + ggplot2::theme(plot.title = ggplot2::element_text(size=18))
+        ggplot2::ggtitle(paste0("Multiome Overlap for ATAC Data\n(By ATAC Majority Vote Cell Type)\n", sample_text)) + ggplot2::theme(plot.title = ggplot2::element_text(size=18), legend.text = ggplot2::element_text(size=10))
       ggplot2::ggsave(filename = paste0(ATAC_output_dir, "Final_Multiome_Overlap_ATAC_UMAP_by_Original_ATAC_Majority_Vote_Cell_Type_Labels.png"), plot = p2, device = "png", width = 8, height = 8, units = "in")
       ArchR::plotPDF(p1,p2, name = "UMAP_multiome_ATAC_with_RNA_labels_and_original_labels_plots", ArchRProj = proj, addDOC = FALSE, width = 5, height = 5)
       # RNA
       sample_count <- length(unique(sc_obj$sample))
       cell_count <- length(sc_obj$cell_name)
-      current_title <- paste0("Multiome Overlap for RNA Data \n (By RNA Majority Vote Cell Type) \n (", sample_count, " Samples, ", cell_count, " Cells)")
+      current_title <- paste0("Multiome Overlap for RNA Data\n(By RNA Majority Vote Cell Type)\n(", sample_count, " Samples, ", cell_count, " Cells)")
       print_UMAP_RNA(sc_obj, file_name = "Final_Multiome_Overlap_RNA_UMAP_by_RNA_Majority_Vote_Cell_Type_Labels.png",
                      group_by_category = "predicted_celltype_majority_vote", output_dir = RNA_output_dir, title = current_title,
                      log_flag = log_flag)

R/process_raw_data.R

@@ -290,7 +290,7 @@ InitialProcessing_RNA <- function(sc_obj, species = "human", output_dir = getwd(
       sc_obj <- Seurat::RunUMAP(sc_obj, reduction = "pca", dims = 1:30, seed.use = get_speedi_seed())
       sample_count <- length(unique(sc_obj$sample))
       cell_count <- length(sc_obj$cell_name)
-      current_title <- paste0("RNA Data Before Integration \n (By Sample) \n (", sample_count, " Samples, ", cell_count, " Cells)")
+      current_title <- paste0("RNA Data Before Integration\n(By Sample)\n(", sample_count, " Samples, ", cell_count, " Cells)")
       print_UMAP_RNA(sc_obj, file_name = "Before_Batch_Correction_RNA_UMAP_by_Sample.png",
                      group_by_category = "sample", output_dir = output_dir, title = current_title,
                      log_flag = log_flag)
@@ -358,10 +358,10 @@ InitialProcessing_ATAC <- function(proj, output_dir = getwd(), exit_with_code =
         sample_text <- paste0("(", num_samples, " Samples, ", num_cells, " Cells)")
       }
       p1 <- ArchR::plotEmbedding(ArchRProj = proj, colorBy = "cellColData", name = "Sample", embedding = "UMAP", force = TRUE, keepAxis = TRUE) +
-        ggplot2::ggtitle(paste0("ATAC Data Before Integration (By Sample) \n ", sample_text)) + ggplot2::theme(plot.title = ggplot2::element_text(size=18))
+        ggplot2::ggtitle(paste0("ATAC Data Before Integration\n(By Sample)\n", sample_text)) + ggplot2::theme(plot.title = ggplot2::element_text(size=18), legend.text = ggplot2::element_text(size=10))
       ggplot2::ggsave(filename = paste0(output_dir, "Before_Batch_Correction_ATAC_UMAP_by_Sample.png"), plot = p1, device = "png", width = 8, height = 8, units = "in")
       p2 <- ArchR::plotEmbedding(ArchRProj = proj, colorBy = "cellColData", name = "TSSEnrichment", embedding = "UMAP", force = TRUE, keepAxis = TRUE) +
-        ggplot2::ggtitle(paste0("ATAC Data Before Integration (By TSS Enrichment) \n ", sample_text)) + ggplot2::theme(plot.title = ggplot2::element_text(size=18))
+        ggplot2::ggtitle(paste0("ATAC Data Before Integration\n(By TSS Enrichment)\n", sample_text)) + ggplot2::theme(plot.title = ggplot2::element_text(size=18), legend.key.size = ggplot2::unit(1, "cm"), legend.text = ggplot2::element_text(size=10))
       ggplot2::ggsave(filename = paste0(output_dir, "Before_Batch_Correction_ATAC_UMAP_by_TSSEnrichment.png"), plot = p2, device = "png", width = 8, height = 8, units = "in")
       ArchR::plotPDF(p1,p2, name = "UMAPs_After_Initial_Processing_Plots", ArchRProj = proj, addDOC = FALSE, width = 5, height = 5)
       print_SPEEDI("Step 4: Complete", log_flag)

R/qc.R

@@ -11,15 +11,15 @@ Create_QC_Output_Prefiltered_RNA <- function(sc_obj, output_dir = getwd(), log_f
   # Normalize paths (in case user provides relative paths)
   output_dir <- normalize_dir_path(output_dir)
   print_SPEEDI("Printing QC plots on pre-filtered data (RNA)", log_flag)
-  p <- Seurat::VlnPlot(sc_obj, features = c("nFeature_RNA"), split.by = "sample", group.by = "sample", raster = FALSE)
+  p <- Seurat::VlnPlot(sc_obj, features = c("nFeature_RNA"), split.by = "sample", group.by = "sample", raster = FALSE) + ggplot2::xlab("Sample")
   ggplot2::ggsave(paste0(output_dir, "pre-filtered_nFeature_violin_plot.png"), plot = p, device = "png", width = 10, height = 10, units = "in")
-  p <- Seurat::VlnPlot(sc_obj, features = c("nCount_RNA"), split.by = "sample", group.by = "sample", raster = FALSE)
+  p <- Seurat::VlnPlot(sc_obj, features = c("nCount_RNA"), split.by = "sample", group.by = "sample", raster = FALSE) + ggplot2::xlab("Sample")
   ggplot2::ggsave(paste0(output_dir, "pre-filtered_nCount_violin_plots.png"), plot = p, device = "png", width = 10, height = 10, units = "in")
-  p <- Seurat::VlnPlot(sc_obj, features = c("percent.mt"), split.by = "sample", group.by = "sample", raster = FALSE)
+  p <- Seurat::VlnPlot(sc_obj, features = c("percent.mt"), split.by = "sample", group.by = "sample", raster = FALSE) + ggplot2::xlab("Sample")
   ggplot2::ggsave(paste0(output_dir, "pre-filtered_percentMT_violin_plots.png"), plot = p, device = "png", width = 10, height = 10, units = "in")
-  p <- Seurat::VlnPlot(sc_obj, features = c("percent.hb"), split.by = "sample", group.by = "sample", raster = FALSE)
+  p <- Seurat::VlnPlot(sc_obj, features = c("percent.hb"), split.by = "sample", group.by = "sample", raster = FALSE) + ggplot2::xlab("Sample")
   ggplot2::ggsave(paste0(output_dir, "pre-filtered_percentHB_violin_plots.png"), plot = p, device = "png", width = 10, height = 10, units = "in")
-  p <- Seurat::VlnPlot(sc_obj, features = c("percent.rp"), split.by = "sample", group.by = "sample", raster = FALSE)
+  p <- Seurat::VlnPlot(sc_obj, features = c("percent.rp"), split.by = "sample", group.by = "sample", raster = FALSE) + ggplot2::xlab("Sample")
   ggplot2::ggsave(paste0(output_dir, "pre-filtered_percentRP_violin_plots.png"), plot = p, device = "png", width = 10, height = 10, units = "in")
   # Create nCount vs nFeature scatter plot for each sample
   individual_samples <- Seurat::SplitObject(sc_obj, split.by = "sample")
@@ -59,13 +59,13 @@ Create_QC_Output_Prefiltered_ATAC <- function(proj, output_dir = getwd(), log_fl
   print_SPEEDI("Printing QC plots on pre-filtered data (ATAC)", log_flag)
   # Plot out TSS Enrichment / Doublet Enrichment / Nucleosome Ratio for each sample to help us decide filtering thresholds
   p1 <- ArchR::plotGroups(ArchRProj = proj, groupBy = "Sample", colorBy = "cellColData", name = "TSSEnrichment", plotAs = "ridges") +
-    ggplot2::ggtitle(paste0("TSS Enrichment \n ", sample_text)) + ggplot2::theme(plot.title = ggplot2::element_text(size=22))
+    ggplot2::ggtitle(paste0("TSS Enrichment\n", sample_text)) + ggplot2::theme(plot.title = ggplot2::element_text(size=18), axis.text=ggplot2::element_text(size=10), axis.title=ggplot2::element_text(size=14))
   ggplot2::ggsave(filename = paste0(output_dir, "pre-filtered_TSSEnrichment.png"), plot = p1, device = "png", width = 8, height = 8, units = "in")
   p2 <- ArchR::plotGroups(ArchRProj = proj, groupBy = "Sample", colorBy = "cellColData", name = "DoubletEnrichment", plotAs = "ridges") +
-    ggplot2::ggtitle(paste0("Doublet Enrichment \n ", sample_text)) + ggplot2::theme(plot.title = ggplot2::element_text(size=22))
+    ggplot2::ggtitle(paste0("Doublet Enrichment\n", sample_text)) + ggplot2::theme(plot.title = ggplot2::element_text(size=18), axis.text=ggplot2::element_text(size=10), axis.title=ggplot2::element_text(size=14))
   ggplot2::ggsave(filename = paste0(output_dir, "pre-filtered_DoubletEnrichment.png"), plot = p2, device = "png", width = 8, height = 8, units = "in")
   p3 <- ArchR::plotGroups(ArchRProj = proj, groupBy = "Sample", colorBy = "cellColData", name = "NucleosomeRatio", plotAs = "ridges") +
-    ggplot2::ggtitle(paste0("Nucleosome Ratio \n ", sample_text)) + ggplot2::theme(plot.title = ggplot2::element_text(size=22))
+    ggplot2::ggtitle(paste0("Nucleosome Ratio\n", sample_text)) + ggplot2::theme(plot.title = ggplot2::element_text(size=18), axis.text=ggplot2::element_text(size=10), axis.title=ggplot2::element_text(size=14))
   ggplot2::ggsave(filename = paste0(output_dir, "pre-filtered_NucleosomeRatio.png"), plot = p3, device = "png", width = 8, height = 8, units = "in")
   ArchR::plotPDF(p1,p2,p3, name = "pre-filtered_QC_metrics_plots", ArchRProj = proj, addDOC = FALSE, width = 7, height = 5)
   gc()

R/utils.R

@@ -234,7 +234,7 @@ print_heatmap_cell_type_proportions_RNA <- function(sc_obj, file_name, output_di
   sample_count <- length(unique(sc_obj$sample))
   cell_count <- length(sc_obj$cell_name)
   output.plot <- pheatmap::pheatmap(voting_cell_type_proportion,  cluster_rows = FALSE, cluster_cols = FALSE, display_numbers = TRUE, number_format = "%.3f", legend = FALSE) +
-    ggplot2::ggtitle(paste0("Cell Type Proportions \n Integrated RNA Data \n (", sample_count, " Samples, ", cell_count, " Cells)")) + ggplot2::theme(plot.title = ggplot2::element_text(size=18))
+    ggplot2::ggtitle(paste0("RNA Data Integration\n(Cell Type Proportions)\n(", sample_count, " Samples, ", cell_count, " Cells)")) + ggplot2::theme(plot.title = ggplot2::element_text(size=18))
   ggplot2::ggsave(filename = paste0(output_dir, file_name), plot = output.plot, device = "png", width = 8, height = 8, units = "in")
   return(TRUE)
 }