Merge branch '6-processCounts' into 'develop'

Resolve "process_count"

See merge request !22

.gitlab-ci.yml

@@ -77,6 +77,13 @@ makeBigWig:
   - singularity run 'docker://bicf/deeptools3.3:2.0.0' bamCoverage -p `nproc` -b ./test_data/bam/small/Q-Y5JA_1M.se.sorted.deduped.bam -o Q-Y5JA_1M.se.bw
   - pytest -m makeBigWig
 
+makeFeatureCounts:
+  stage: unit
+  script:
+  - singularity run 'docker://bicf/subread2:2.0.0' featureCounts -R SAM -p -G /project/BICF/BICF_Core/shared/gudmap/references/GRCh38.p12.v31/genome.fna -T `nproc` -s 1 -a /project/BICF/BICF_Core/shared/gudmap/references/GRCh38.p12.v31/genome.gtf -o Q-Y5JA_1M.se.featureCounts -g 'gene_name' --primary --ignoreDup -B ./test_data/bam/small/Q-Y5JA_1M.se.sorted.deduped.bam 
+  - singularity run 'docker://bicf/subread2:2.0.0' Rscript ./workflow/scripts/calculateTPM.R --count Q-Y5JA_1M.se.featureCounts
+  - pytest -m makeFeatureCounts
+
 fastqc:
   stage: unit
   script:
@@ -92,11 +99,13 @@ inferMetadata:
 integration_se:
   stage: integration
   script:
+  - ulimit -u 16384
   - nextflow run ./workflow/rna-seq.nf --deriva ./test_data/auth/credential.json --bdbag ./test_data/auth/cookies.txt --repRID 16-1ZX4
 
 integration_pe:
   stage: integration
   script:
+  - ulimit -u 16384
   - nextflow run ./workflow/rna-seq.nf --deriva ./test_data/auth/credential.json --bdbag ./test_data/auth/cookies.txt --repRID Q-Y5JA -with-dag dag.png
   artifacts:
     name: "$CI_JOB_NAME"

workflow/nextflow.config

@@ -41,6 +41,9 @@ process {
   withName:inferMetadata{
     container = 'bicf/rseqc3.0:2.0.0'
   }
+  withName: makeFeatureCounts {
+    container = 'bicf/subread2:2.0.0'
+  }
 }
 
 trace {

workflow/scripts/calculateTPM.R

+gc()
+library(optparse)
+
+option_list=list(
+  make_option("--count",action="store",type='character',help="Count File")
+)
+opt=parse_args(OptionParser(option_list=option_list))
+rm(option_list)
+
+if (!("count" %in% names(opt))){
+  stop("No count file passed, exiting.")
+} else if (!file.exists(opt$count)) {
+  stop("Count file doesn't exist, exiting.")
+}
+
+repRID <- basename(gsub(".featureCounts","",opt$count))
+
+count <- read.delim(opt$count, comment.char="#") # if featureCounts file changes structure, be sure to update count and Length columns below
+colnames(count)[7] <- "count"
+
+rpk <- count$count/count$Length/1000
+
+scale <- sum(rpk)/1000000
+
+tpm <- rpk/scale
+
+output <- cbind(count,tpm)
+colnames(output)[7] <- "count"
+
+write.table(output,file=paste0(repRID,".countTable.csv"),sep=",",row.names=FALSE,quote=FALSE)

workflow/tests/test_makeFeatureCounts.py

+#!/usr/bin/env python3
+
+import pytest
+import pandas as pd
+import os
+import utils
+
+data_output_path = os.path.dirname(os.path.abspath(__file__)) + \
+	'/../../'
+
+
+@pytest.mark.makeFeatureCounts
+def test_makeFeatureCounts():
+	assert os.path.exists(os.path.join(data_output_path, 'Q-Y5JA_1M.se.featureCounts'))
+	assert os.path.exists(os.path.join(data_output_path, 'Q-Y5JA_1M.se.countTable.csv'))