nf-core_modules/modules/star/genomegenerate/main.nf

68 lines
2.4 KiB
Text
Raw Normal View History

2020-09-10 15:45:11 +00:00
// Import generic module functions
include { initOptions; saveFiles; getSoftwareName } from './functions'
2020-10-14 17:29:50 +00:00
params.options = [:]
options = initOptions(params.options)
2020-10-14 17:29:50 +00:00
2020-09-10 15:45:11 +00:00
process STAR_GENOMEGENERATE {
tag "$fasta"
label 'process_high'
publishDir "${params.outdir}",
mode: params.publish_dir_mode,
saveAs: { filename -> saveFiles(filename:filename, options:params.options, publish_dir:'index', meta:[:], publish_by_meta:[]) }
2020-09-10 15:45:11 +00:00
2020-10-14 17:29:50 +00:00
// Note: 2.7X indices incompatible with AWS iGenomes.
conda (params.enable_conda ? "bioconda::star=2.7.9a bioconda::samtools=1.13 conda-forge::gawk=5.1.0" : null)
if (workflow.containerEngine == 'singularity' && !params.singularity_pull_docker_container) {
container "https://depot.galaxyproject.org/singularity/mulled-v2-1fa26d1ce03c295fe2fdcf85831a92fbcbd7e8c2:a7908dfb0485a80ca94e4d17b0ac991532e4e989-0"
} else {
container "quay.io/biocontainers/mulled-v2-1fa26d1ce03c295fe2fdcf85831a92fbcbd7e8c2:a7908dfb0485a80ca94e4d17b0ac991532e4e989-0"
}
2020-09-10 15:45:11 +00:00
input:
path fasta
path gtf
output:
path "star" , emit: index
path "*.version.txt", emit: version
script:
def software = getSoftwareName(task.process)
def memory = task.memory ? "--limitGenomeGenerateRAM ${task.memory.toBytes() - 100000000}" : ''
def args = options.args.tokenize()
if (args.contains('--genomeSAindexNbases')) {
"""
mkdir star
STAR \\
--runMode genomeGenerate \\
--genomeDir star/ \\
--genomeFastaFiles $fasta \\
--sjdbGTFfile $gtf \\
--runThreadN $task.cpus \\
$memory \\
$options.args
STAR --version | sed -e "s/STAR_//g" > ${software}.version.txt
"""
} else {
"""
samtools faidx $fasta
NUM_BASES=`gawk '{sum = sum + \$2}END{if ((log(sum)/log(2))/2 - 1 > 14) {printf "%.0f", 14} else {printf "%.0f", (log(sum)/log(2))/2 - 1}}' ${fasta}.fai`
mkdir star
STAR \\
--runMode genomeGenerate \\
--genomeDir star/ \\
--genomeFastaFiles $fasta \\
--sjdbGTFfile $gtf \\
--runThreadN $task.cpus \\
--genomeSAindexNbases \$NUM_BASES \\
$memory \\
$options.args
STAR --version | sed -e "s/STAR_//g" > ${software}.version.txt
"""
}
2020-09-10 15:45:11 +00:00
}