fixed fastqc summary

dge_workflow/dge_utils.sh

@@ -6,6 +6,10 @@ source <(curl https://raw.githubusercontent.com/holgerbrandl/datautils/v1.14/bas
 
 ## enable snippet spinning
 source <(curl https://raw.githubusercontent.com/holgerbrandl/datautils/v1.14/R/spinr/spin_utils.sh 2>&1 2>/dev/null)
+export PATH=/home/brandl/bin/spinr:$PATH
+
+source <(curl https://raw.githubusercontent.com/holgerbrandl/datautils/master/R/rendr/rendr_utils.sh 2>&1 2>/dev/null)
+export PATH=/home/brandl/bin/rendr:$PATH
 
 
 ## define common binaries
@@ -17,7 +21,6 @@ export PATH=/sw/apps/python/current/bin:$PATH
 export PATH=/home/brandl/bin/deepTools/bin:$PATH
 export PATH=/projects/bioinfo/holger/bin/FastQC_0.11.2:$PATH
 export PATH=/projects/bioinfo/holger/bin/bedtools-2.23.0/bin/:$PATH
-export PATH=/home/brandl/bin/spinr:$PATH
 export PATH=/home/brandl/bin/subread-1.4.6-p3-Linux-x86_64/bin:$PATH
 
 #export PATH=/home/brandl/bin/STAR/STAR-STAR_2.4.1d/source:$PATH
@@ -75,7 +78,7 @@ done
 wait4jobs .fastqc_jobs
 
 
-spin.R ${NGS_TOOLS}/dge_workflow/fastqc_summary.R $outputDir
+rend.R ${NGS_TOOLS}/dge_workflow/fastqc_summary.R $outputDir
 
 mailme "$project: fastqc done in $(pwd)"
 

dge_workflow/fastqc_summary.R

@@ -4,8 +4,8 @@
 
 ## Note This script is supposed to be knitr::spin'ed
 
-devtools::source_url("https://raw.githubusercontent.com/holgerbrandl/datautils/v1.9/R/core_commons.R")
-devtools::source_url("https://raw.githubusercontent.com/holgerbrandl/datautils/v1.9/R/ggplot_commons.R")
+devtools::source_url("https://raw.githubusercontent.com/holgerbrandl/datautils/v1.14/R/core_commons.R")
+devtools::source_url("https://raw.githubusercontent.com/holgerbrandl/datautils/v1.14/R/ggplot_commons.R")
 
 ## can we access variables from the parent spin.R process?
 #echo("rscript is ", r_script)
@@ -20,6 +20,7 @@ if(length(argv) != 1){
 }
 
 baseDir=argv[1]
+#baseDir=normalizePath(".")
 
 
 if(is.na(file.info(baseDir)$isdir)){
@@ -50,7 +51,7 @@ readCount <- function(statsFile){
 
 readCounts <- fastqDataFiles %>% ldply(readCount) # %>% print_head()
 
-require.auto(scales)
+
 #+ fig.width=12, fig.height=round(nrow(readCounts)/2)
 ggplot(readCounts, aes(run, num_reads)) + geom_bar(stat="identity") + coord_flip() + scale_y_continuous(labels=comma) + ggtitle("read counts")
 
@@ -92,7 +93,6 @@ dupLevels <- fastqDataFiles %>% ldply(function(statsFile){
    )
 })
 
-require.auto(scales)
 #+ fig.width=12, fig.height=round(nrow(dupLevels)/2)
 ggplot(dupLevels, aes(run, dedup_proportion)) + geom_bar(stat="identity") + coord_flip() + scale_y_continuous(labels=percent) + ggtitle("unique_reads/total_reads") + ylim(0,1)
 
@@ -110,7 +110,7 @@ readBaseQualDist <- function(statsFile){
 #    echo("reading", statsFile)
 
     baseStats <- read.delim(pipe(
-            paste(get_zip_pipe(statsFile, "fastqc_data.txt"), " | grep -A60 -F '>>Per base sequence quality' | grep -B100 -F '>>END_MODULE' | head -n-1 | tail -n+2 | tr '#' ' '")
+            paste(get_zip_pipe(statsFile, "fastqc_data.txt"), " | grep -A200 -F '>>Per base sequence quality' | grep -B200 -F 'Per tile sequence quality' | head -n-2 | tail -n+2 | tr '#' ' '")
         )) %>% mutate(
             run=trim_ext(basename(statsFile), c(".zip"))
         )
@@ -119,14 +119,17 @@ readBaseQualDist <- function(statsFile){
 }
 
 baseQualities <- fastqDataFiles %>%  ldply(readBaseQualDist)
-statsFile="fastqc/L8038_Track-21511_R1.trim_fastqc.zip"
+#statsFile="fastqc/L8038_Track-21511_R1.trim_fastqc.zip"
 #with(baseQualities, as.data.frame(table(run)))
 
 
 #+ fig.widthh=20
-baseQualities %>% ggplot(aes(reorder(Base, base_order), Mean, group=run, color=run)) + geom_line() + scale_y_continuous(limits=c(2, 40))
+seqQualPlot <- baseQualities %>% ggplot(aes(reorder(Base, base_order), Mean, group=run, color=run)) + geom_line() + scale_y_continuous(limits=c(2, 40))
 
+## just show color legend if not too many samples
+if(unlen(baseQualities$run) > 20) seqQualPlot <- seqQualPlot + guides(color=F)
 
+seqQualPlot
 
 #' # Qualities per run including variance