added new tech-rep-renaming

dge_workflow/lsf_rna_seq.sh

@@ -191,7 +191,10 @@ cp -r . $tmpDbDir
 
 ## todo remove this hack
 genome=$(echo $gtfFile | cut -f7 -d'/'); echo "genome is $genome"
+
+## make sure to use temp-r to avoid file locking problems
 R_LIBS=/tmp/r_index
+
 echo '
 require(cummeRbund)
 dbDir=commandArgs(T)[1]

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
+
+
+1) evaluate if trimmoatic is the better trimmer (with respect to cutadapt)
+- also consider to use contamination list from fastqc for trimming (see https://www.biostars.org/p/15753/)
\ No newline at end of file

misc/snippets.sh

+#########################################################################################################################
+#### Trimmomatic
+
+zcat  $fastqFile | head -n 400000 > test.fastq
+gzip test.fastq
+
+export TRIMMOMATIC_HOME=/projects/bioinfo/holger/bin/Trimmomatic-0.32
+
+# do the filtering
+mcdir $baseDir/trimmed
+
+for fastqFile in $baseDir/treps_pooled/*fastq.gz ; do
+    # DEBUG fastqFile=/projects/bioinfo/holger/projects/helin/dog/treps_pooled/dog_big_cyst_rep1.fastq.gz
+    # DEBUG fastqFile=test.fastq.gz
+
+    caFastq=$(basename $fastqFile .fastq.gz)_tm.fastq.gz
+    echo "cutadapting $caFastq into $caFastq"
+
+    cmd="java -Xmx2g -jar $TRIMMOMATIC_HOME/trimmomatic-0.32.jar SE -threads 1 -phred33 $fastqFile $caFastq ILLUMINACLIP:$TRIMMOMATIC_HOME/adapters/TruSeq3-SE.fa:2:30:10 LEADING:3 TRAILING:3 SLIDINGWINDOW:4:15 MINLEN:36"
+    eval $cmd
+    mysub "$project__ca__$caFastq" "cutadapt -m 20 -q 25 -o $caFastq $fastqFile > $caFastq.ca.log"  -q long  | joblist .tmjobs
+done
+wait4jobs .tmjobs
+
+
+dge_fastqc -o $baseDir/fastqc_tm $(ls $baseDir/trimmed/*fastq.gz)
+
+mailme "trimmomatic done"
+
+