From 11a742a37c0a68c6d70361db30fc085817ff212d Mon Sep 17 00:00:00 2001
From: "Gervaise H. Henry" <gervaise.henry@utsouthwestern.edu>
Date: Tue, 4 Aug 2020 09:36:24 -0500
Subject: [PATCH] Add new metadata extraction to ci
---
.gitlab-ci.yml | 3 +++
workflow/rna-seq.nf | 17 +++++++++--------
2 files changed, 12 insertions(+), 8 deletions(-)
diff --git a/.gitlab-ci.yml b/.gitlab-ci.yml
index 29f4925..6264f76 100644
--- a/.gitlab-ci.yml
+++ b/.gitlab-ci.yml
@@ -67,6 +67,8 @@ trimData:
script:
- singularity run 'docker://bicf/trimgalore:1.1' trim_galore --gzip -q 25 --illumina --length 35 --basename Q-Y5F6_1M.se -j 20 ./test_data/fastq/small/Q-Y5F6_1M.R1.fastq.gz
- singularity run 'docker://bicf/trimgalore:1.1' trim_galore --gzip -q 25 --illumina --length 35 --paired --basename Q-Y5F6_1M.pe -j 20 ./test_data/fastq/small/Q-Y5F6_1M.R1.fastq.gz ./test_data/fastq/small/Q-Y5F6_1M.R2.fastq.gz
+ - readLengthSE=$(zcat *_trimmed.fq.gz | awk '{if(NR%4==2) print length($1)}' | sort -n | awk '{a[NR]=$0}END{print(NR%2==1)?a[int(NR/2)+1]:(a[NR/2]+a[NR/2+1])/2}')
+ - readLengthPE=$(zcat *_1.fq.gz | awk '{if(NR%4==2) print length($1)}' | sort -n | awk '{a[NR]=$0}END{print(NR%2==1)?a[int(NR/2)+1]:(a[NR/2]+a[NR/2+1])/2}')
- pytest -m trimData
downsampleData:
@@ -105,6 +107,7 @@ countData:
script:
- singularity run 'docker://bicf/subread2:2.0.0' featureCounts -T 20 -a /project/BICF/BICF_Core/shared/gudmap/references/GRCh38.p12.v31/genome.gtf -G /project/BICF/BICF_Core/shared/gudmap/references/GRCh38.p12.v31/genome.fna -g 'gene_name' -o Q-Y5F6_1M.se.featureCounts -s 1 -R SAM --primary --ignoreDup ./test_data/bam/small/Q-Y5F6_1M.se.sorted.deduped.bam
- singularity run 'docker://bicf/subread2:2.0.0' Rscript ./workflow/scripts/calculateTPM.R --count ./test_data/counts/small/Q-Y5F6_1M.se.featureCounts
+ - assignedReads=grep -m 1 'Assigned' *.countData.summary | grep -oe '\([0-9.]*\)'
- pytest -m makeFeatureCounts
makeBigWig:
diff --git a/workflow/rna-seq.nf b/workflow/rna-seq.nf
index a8a63aa..4be9f4f 100644
--- a/workflow/rna-seq.nf
+++ b/workflow/rna-seq.nf
@@ -298,7 +298,7 @@ process trimData {
hostname > ${repRID}.trimData.log
ulimit -a >> ${repRID}.trimData.log
- # trim fastq's using trim_galore
+ # trim fastq's using trim_galore and extract median read length
echo -e "LOG: trimming ${ends}" >> ${repRID}.trimData.log
if [ "${ends}" == "se" ]
then
@@ -875,7 +875,8 @@ process countData {
featureCounts -T `nproc` -a ./genome.gtf -G ./genome.fna -g 'gene_name' -o ${repRID}.countData -s \${stranding} -p -B -R SAM --primary --ignoreDup ${repRID}.sorted.deduped.bam
fi
echo -e "LOG: counted" >> ${repRID}.countData.log
-
+
+ # extract assigned reads
assignedReads=grep -m 1 'Assigned' *.countData.summary | grep -oe '\([0-9.]*\)'
echo -e \${assignedReads} > assignedReads.csv
echo -e "LOG: assigned reads: "\${assignedReads} >> ${repRID}.countData.log
@@ -887,6 +888,12 @@ process countData {
"""
}
+// Extract number of assigned reads metadata into channel
+assignedReadsInfer = Channel.create()
+inferMetadata_assignedReads.splitCsv(sep: ",", header: false).separate(
+ assignedReads
+)
+
/*
*fastqc: run fastqc on untrimmed fastq's
*/
@@ -911,12 +918,6 @@ process fastqc {
"""
}
-// Extract number of assigned reads metadata into channel
-assignedReadsInfer = Channel.create()
-inferMetadata_assignedReads.splitCsv(sep: ",", header: false).separate(
- assignedReads
-)
-
/*
*dataQC: calculate transcript integrity numbers (TIN) and bin as well as calculate innerdistance of PE replicates
*/
--
GitLab