cont. rna-seq workflow

dge_workflow/fastqc_summary.R

+#!/usr/bin/env Rscript
 #' # FastQC Summary for `r getwd()`
 ##+ echo=FALSE, message=FALSE
 
@@ -23,7 +24,7 @@ baseDir=argv[1]
 
 
 if(is.na(file.info(baseDir)$isdir)){
-    stop(paste("movie directory does not exist"))
+    stop(paste("directory '", baseDir,"'does not exist"))
 }
 
 #baseDir="fastqc"

dge_workflow/lsf_rna_seq.sh

@@ -48,7 +48,6 @@ mailme "fastqc done for $outputDir"dge_tophat_se
 
 ziprm fastqc_logs fastqc__*
 
-# todo create some summary report here
 spin.R $DGE_HOME/fastqc_summary.R $outputDir
 
 }
@@ -65,10 +64,13 @@ for fastqFile in $* ; do
     echo "cutadapting $caFastq into $caFastq"
 
     #todo use a more specific trimming model (trim just correct part on each side without using reverse complements
-    mysub "$project__ca__$caFastq" "cutadapt -b file:$Ill_ADAPTERS -m 20 -q 25 -o $caFastq $fastqFile > $caFastq.ca.log"  -q long  | joblist .cajobs
+    mysub "$project__ca__$(basename $fastqFile .fastq.gz)" "cutadapt -b file:$Ill_ADAPTERS -m 20 -q 25 -o $caFastq $fastqFile"  -q long  | joblist .cajobs
 done
 
 wait4jobs .cajobs
+ziprm cutadapt_logs $project__ca__*
+
+## todo do a small report here about what has been trimmed away and why
 
 }
 export -f dge_cutadapt
@@ -91,8 +93,6 @@ shift $(($OPTIND - 1))
 local fastqFiles=$*
 
 # IGENOME=/projects/bioinfo/igenomes/Mus_musculus/Ensembl/GRCm38
-echo fastq $fastqFiles
-echo igenomes "$IGENOME"
 
 if [ -z "$IGENOME" ] || [ -z "$fastqFiles" ];
      then echo "Usage: dge_tophat_se -i <path to igenome> [<fastq.gz file>]+" >&2 ; return;
@@ -143,9 +143,6 @@ spin.R $DGE_HOME/bam_qc.R .
 }
 export -f dge_tophat_se
 
-# dge_tophat_se
-# dge_tophat_se -i
-
 
 dge_cuffdiff(){
 

dge_workflow/todo.txt

 - cuffdbs change dramatically in size if gtf is provided when building them, but what impact does it have on the results
+
+- Also try to remove RNA PCR Primer enrichment. Currently we just remove index and universal adapter

misc/spin.R

@@ -51,10 +51,10 @@ commandArgs <- function(trailingOnly = FALSE){ return(as.character(spin_opts$scr
 
 spin(r_script, knit=F)
 
-mdScript <- str_replace(r_script, "[.]R$", ".Rmd")
+rmdScript <- str_replace(r_script, "[.]R$", ".Rmd")
 
-system(paste("mv", mdScript, "tmp.Rmd"))
-system(paste("cat tmp.Rmd | grep -Ev '^#+$' | grep -Fv '#!/usr/bin/env Rscript' >", basename(mdScript)))
+system(paste("mv", rmdScript, "tmp.Rmd"))
+system(paste("cat tmp.Rmd | grep -Ev '^#+$' | grep -Fv '#!/usr/bin/env Rscript' >", basename(rmdScript)))
 file.remove("tmp.Rmd")
 
 cssHeader='
@@ -67,14 +67,17 @@ cssHeader='
 
 ## custom title http://stackoverflow.com/questions/14124022/setting-html-meta-elements-with-knitr
 opts_chunk$set(
-    cache = spin_opts$"-c",
-    message= spin_opts$"-m",
-    warning= spin_opts$"-w",
-    echo= spin_opts$"-e",
+    cache = spin_opts$c,
+    message= spin_opts$m,
+    warning= spin_opts$w,
+    echo= spin_opts$e,
     fig.width=15,
     width=200
 )
 
-knit2html(basename(mdScript), header=cssHeader)
+knit2html(basename(rmdScript), header=cssHeader)
+
+## also remove the .md and the .Rmd files
+file.remove(basename(rmdScript))
+file.remove(basename(str_replace(r_script, "[.]R$", ".md")))
 
-file.remove(basename(mdScript))