fix: experiment count QC plots

workflow/rules/qc_report.smk

@@ -83,8 +83,9 @@ rule qc_report_count:
         statistics_all_single="results/experiments/{project}/statistic/statistic_assigned_counts_single_{assignment}_{config}.tsv",
         statistics_all_oligo_cor_all="results/experiments/{project}/statistic/assigned_counts/{assignment}/{config}/{condition}_correlation.tsv",
         statistics_all_oligo_cor_thresh="results/experiments/{project}/statistic/assigned_counts/{assignment}/{config}/{condition}_correlation_minThreshold.tsv",
-        counts_per_oligo_dna="results/experiments/{project}/statistic/barcode/assigned_counts/{assignment}/{condition}_{config}_DNA_perBarcode.png",
-        counts_per_oligo_rna="results/experiments/{project}/statistic/barcode/assigned_counts/{assignment}/{condition}_{config}_RNA_perBarcode.png",
+        counts_per_oligo="results/experiments/{project}/statistic/assigned_counts/{assignment}/{config}/{condition}_barcodesPerInsert.png",
+        counts_per_bc_dna="results/experiments/{project}/statistic/barcode/assigned_counts/{assignment}/{condition}_{config}_DNA_perBarcode.png",
+        counts_per_bc_rna="results/experiments/{project}/statistic/barcode/assigned_counts/{assignment}/{condition}_{config}_RNA_perBarcode.png",
         activity_thresh="results/experiments/{project}/statistic/assigned_counts/{assignment}/{config}/{condition}_group_barcodesPerInsert_box.png",
         activity_all="results/experiments/{project}/statistic/assigned_counts/{assignment}/{config}/{condition}_group_barcodesPerInsert_box_minThreshold.png",
         dna_over_rna="results/experiments/{project}/statistic/assigned_counts/{assignment}/{config}/{condition}_dna_vs_rna.png",
@@ -121,8 +122,9 @@ rule qc_report_count:
             -P "ratio_oligo_coor_plot:{input.ratio_oligo_coor_plot}" \
             -P "ratio_oligo_min_thre_plot:{input.ratio_oligo_min_thre_plot}" \
             -P "statistics_all_merged:{input.statistics_all_merged}" \
-            -P "counts_per_oligo_dna:{input.counts_per_oligo_dna}" \
-            -P "counts_per_oligo_rna:{input.counts_per_oligo_rna}" \
+            -P "counts_per_oligo:{input.counts_per_oligo}" \
+            -P "counts_per_bc_dna:{input.counts_per_bc_dna}" \
+            -P "counts_per_bc_rna:{input.counts_per_bc_rna}" \
             -P "statistics_all_single:{input.statistics_all_single}" \
             -P "activity_all:{input.activity_all}" \
             -P "activity_thresh:{input.activity_thresh}" \

workflow/rules/statistic/correlation.smk

@@ -354,6 +354,7 @@ rule statistic_correlation_hist_box_plots:
         ),
         script=getScript("count/plot_perInsertCounts_stats.R"),
     output:
+        "results/experiments/{project}/statistic/assigned_counts/{assignment}/{config}/{condition}_barcodesPerInsert.png",
         "results/experiments/{project}/statistic/assigned_counts/{assignment}/{config}/{condition}_group_barcodesPerInsert_box.png",
         "results/experiments/{project}/statistic/assigned_counts/{assignment}/{config}/{condition}_group_barcodesPerInsert_box_minThreshold.png",
         "results/experiments/{project}/statistic/assigned_counts/{assignment}/{config}/{condition}_dna_vs_rna.png",

workflow/scripts/count/plot_perInsertCounts_stats.R

@@ -195,12 +195,17 @@ for (n in 1:(data %>% nrow())) {
   assigned_counts <- all %>% filter(replicate == rep)
 
   # Histograms
-  intercept <- median(assigned_counts$n_bc)
+  x_lim_n_bc <- min(max(assigned_counts$n_bc), 300)
+  intercept_median <- median(assigned_counts$n_bc)
+  intercept_mean <- mean(assigned_counts$n_bc)
   hist_plot_list[[n]] <-
     ggplot(assigned_counts, aes(x = n_bc)) +
-    geom_histogram(bins = 300) +
-    geom_vline(xintercept = intercept, colour = "red") +
-    xlim(0, 300) +
+    geom_histogram(bins = x_lim_n_bc) +
+    geom_vline(xintercept = intercept_median, colour = "red") +
+    geom_vline(xintercept = intercept_mean, colour = "blue") +
+    xlim(0, x_lim_n_bc) +
+    ylab("Frequency") +
+    xlab("Barcodes per oligo") +
     ggtitle(paste("replicate", rep, sep = " "))
 
   # Boxplots
@@ -214,22 +219,22 @@ for (n in 1:(data %>% nrow())) {
 }
 
 hist_plot <- do.call("plot_grid", c(hist_plot_list))
-ggsave(sprintf("%s_barcodesPerInsert.png", outdir), hist_plot, width = 30, height = 30)
+ggsave(sprintf("%s_barcodesPerInsert.png", outdir), hist_plot, dpi = 300, type = "cairo")
 
 
 
 box_plot_insert <- do.call("plot_grid", c(box_plot_insert_list))
 ggsave(sprintf("%s_group_barcodesPerInsert_box.png", outdir),
   box_plot_insert,
-  width = 30, height = 30
+  dpi = 300, type = "cairo"
 )
 
 box_plot_insert_thresh <-
   do.call("plot_grid", c(box_plot_insert_thresh_list))
 ggsave(
   sprintf("%s_group_barcodesPerInsert_box_minThreshold.png", outdir),
   box_plot_insert_thresh,
-  width = 30, height = 30
+  dpi = 300, type = "cairo"
 )
 
 print("Script done")

workflow/scripts/report/qc_report_count.qmd

@@ -229,20 +229,33 @@ df_single.style.hide(axis="index").format( precision=2)
 
 # Number of barcodes per oligo
 
-Histogramm of number of barcodes per oligo. Median is blue, mean is red.
+ Median is red, mean is blue.
+
+
+# Histograms barcodes per oligo, counts per barcode
+
+Histogramm of number of barcodes per oligo and the number of counts per barcode devidied by DNA and RNA. Median is red, mean is blue.
+
 
 ::: {.panel-tabset}
 
-# DNA
+# Number of barcodes per oligo
+```{python}
+#| echo: false
+display(Image(f"{workdir}/{counts_per_oligo}")) 
+```
+
+
+# Number of counts per DNA barcode
 ```{python}
 #| echo: false
-display(Image(f"{workdir}/{counts_per_oligo_dna}")) 
+display(Image(f"{workdir}/{counts_per_bc_dna}")) 
 ```
 
-# RNA
+# Number of counts per DNA barcode
 ```{python}
 #| echo: false
-display(Image(f"{workdir}/{counts_per_oligo_rna}")) 
+display(Image(f"{workdir}/{counts_per_bc_rna}")) 
 ```
 
 :::