default parameters added for bowtie2, hisat2 and stringtie to ease ad…

…apting new versions

iLoop_RNAseq_pipeline/defaults/RNAseq_pipeline_defaults.txt

@@ -2,7 +2,10 @@ Note,You can use this file as template. Modify it and leave it under your home f
 email,youremail@here . com
 project,projectid
 RNAseq_env,python_environment_for_RNAseq
+bowtie2_path,/home/projects/cu_10010/programs/bowtie2-2.3.2
+hisat2_path,/home/projects/cu_10010/programs/hisat2-2.1.0
 hisat2_options,-q --no-spliced-alignment --rna-strandness RF --dta
+stringtie_path,/home/projects/cu_10010/programs/stringtie-1.3.3b.Linux_x86_64
 stringtie_options,-B -e
 tophat2_options,
 cufflinks_options,--library-type fr-firststrand --max-bundle-frags 25000000

iLoop_RNAseq_pipeline/master_qsub.py

@@ -107,10 +107,10 @@ def mapandlink_jobs(project_path, sample, reads, defaults, ref, jobs, ppn='8', w
                    .replace('EMAILADDRESS', defaults['email'])]
 
     jobstr += ['''# Load modules needed by myapplication.x
-module load ngs tools samtools/1.2 bowtie2/2.2.5 cufflinks/2.2.1
-# hisat2 and stringtie modules are not kept up-to-date, so use the local ones:
-PATH=$PATH:/home/projects/cu_10010/programs/hisat2-2.0.4:/home/projects/cu_10010/programs/stringtie-1.2.3.Linux_x86_64
-export PATH''']
+module load ngs tools samtools/1.2 cufflinks/2.2.1
+# bowtie2, hisat2 and stringtie modules are not kept up-to-date, so use the local ones:
+PATH=$PATH:{bowtie2_path}:{hisat2_path}:{stringtie_path}
+export PATH'''.format(bowtie2_path=defaults['bowtie2_path'], hisat2_path=defaults['hisat2_path'], stringtie_path=defaults['stringtie_path'])]
 
     jobstr += ['export hisat2_genomic_indexes={}'.format(ref['hisat2_genomic_indexes'])]
     if (map_to_mask) and (not os.path.exists(join_path(project_path, 'map_to_mask', sample))):

iLoop_RNAseq_pipeline/scripts/edgeR_script.r

@@ -71,7 +71,7 @@ y <- DGEList(counts=counts, group=group)
 ## This is a manual filter to keep only Features that have a CPM>0.5 in at least two sample. Features with less are
 ## considered not expressed and won't be included in comparisons.
 ## CPM bound depends very much on the read/genome size. A CPM corresponding to 10 counts should be aimed for. Ideally,
-## this shouldn't not be a static value. Since we usually get generous amount of reads, it will be set to 0.5 for now.
+## this should not be a static value. Since we usually get generous amount of reads, it will be set to 0.5 for now.
 ## This may not apply in all conditions, so a switch as an argument might be useful later on.
 keep <- rowSums(cpm(y)>opt$filter) >= 2
 y <- y[keep, , keep.lib.sizes=FALSE]