module: bwa/aln (#624)
* Specify more guidelines on input channels * Linting * Updates based on code review * Update README.md * Fix broken sentence * Add bwa/aln module * Also output reads as required with SAI * fix container paths * Sync bwa version samese/sampe * Apply suggestions from code review Co-authored-by: Harshil Patel <drpatelh@users.noreply.github.com>jfnavarro-patch-1
James A. Fellows Yates
3 years ago
committed by
GitHub
parent
a21cc95c69
commit
45dee96bdf
@ -0,0 +1,68 @@
|
||||
//
|
||||
// Utility functions used in nf-core DSL2 module files
|
||||
//
|
||||
|
||||
//
|
||||
// Extract name of software tool from process name using $task.process
|
||||
//
|
||||
def getSoftwareName(task_process) {
|
||||
return task_process.tokenize(':')[-1].tokenize('_')[0].toLowerCase()
|
||||
}
|
||||
|
||||
//
|
||||
// Function to initialise default values and to generate a Groovy Map of available options for nf-core modules
|
||||
//
|
||||
def initOptions(Map args) {
|
||||
def Map options = [:]
|
||||
options.args = args.args ?: ''
|
||||
options.args2 = args.args2 ?: ''
|
||||
options.args3 = args.args3 ?: ''
|
||||
options.publish_by_meta = args.publish_by_meta ?: []
|
||||
options.publish_dir = args.publish_dir ?: ''
|
||||
options.publish_files = args.publish_files
|
||||
options.suffix = args.suffix ?: ''
|
||||
return options
|
||||
}
|
||||
|
||||
//
|
||||
// Tidy up and join elements of a list to return a path string
|
||||
//
|
||||
def getPathFromList(path_list) {
|
||||
def paths = path_list.findAll { item -> !item?.trim().isEmpty() } // Remove empty entries
|
||||
paths = paths.collect { it.trim().replaceAll("^[/]+|[/]+\$", "") } // Trim whitespace and trailing slashes
|
||||
return paths.join('/')
|
||||
}
|
||||
|
||||
//
|
||||
// Function to save/publish module results
|
||||
//
|
||||
def saveFiles(Map args) {
|
||||
if (!args.filename.endsWith('.version.txt')) {
|
||||
def ioptions = initOptions(args.options)
|
||||
def path_list = [ ioptions.publish_dir ?: args.publish_dir ]
|
||||
if (ioptions.publish_by_meta) {
|
||||
def key_list = ioptions.publish_by_meta instanceof List ? ioptions.publish_by_meta : args.publish_by_meta
|
||||
for (key in key_list) {
|
||||
if (args.meta && key instanceof String) {
|
||||
def path = key
|
||||
if (args.meta.containsKey(key)) {
|
||||
path = args.meta[key] instanceof Boolean ? "${key}_${args.meta[key]}".toString() : args.meta[key]
|
||||
}
|
||||
path = path instanceof String ? path : ''
|
||||
path_list.add(path)
|
||||
}
|
||||
}
|
||||
}
|
||||
if (ioptions.publish_files instanceof Map) {
|
||||
for (ext in ioptions.publish_files) {
|
||||
if (args.filename.endsWith(ext.key)) {
|
||||
def ext_list = path_list.collect()
|
||||
ext_list.add(ext.value)
|
||||
return "${getPathFromList(ext_list)}/$args.filename"
|
||||
}
|
||||
}
|
||||
} else if (ioptions.publish_files == null) {
|
||||
return "${getPathFromList(path_list)}/$args.filename"
|
||||
}
|
||||
}
|
||||
}
|
||||
@ -0,0 +1,67 @@
|
||||
// Import generic module functions
|
||||
include { initOptions; saveFiles; getSoftwareName } from './functions'
|
||||
|
||||
params.options = [:]
|
||||
options = initOptions(params.options)
|
||||
|
||||
process BWA_ALN {
|
||||
tag "$meta.id"
|
||||
label 'process_medium'
|
||||
publishDir "${params.outdir}",
|
||||
mode: params.publish_dir_mode,
|
||||
saveAs: { filename -> saveFiles(filename:filename, options:params.options, publish_dir:getSoftwareName(task.process), meta:meta, publish_by_meta:['id']) }
|
||||
|
||||
conda (params.enable_conda ? "bioconda::bwa=0.7.17" : null)
|
||||
if (workflow.containerEngine == 'singularity' && !params.singularity_pull_docker_container) {
|
||||
container "https://depot.galaxyproject.org/singularity/bwa:0.7.17--h5bf99c6_8"
|
||||
} else {
|
||||
container "quay.io/biocontainers/bwa:0.7.17--h5bf99c6_8"
|
||||
}
|
||||
|
||||
input:
|
||||
tuple val(meta), path(reads)
|
||||
path index
|
||||
|
||||
output:
|
||||
tuple val(meta), path(reads), path("*.sai"), emit: sai
|
||||
path "*.version.txt" , emit: version
|
||||
|
||||
script:
|
||||
def software = getSoftwareName(task.process)
|
||||
def prefix = options.suffix ? "${meta.id}${options.suffix}" : "${meta.id}"
|
||||
|
||||
if (meta.single_end) {
|
||||
"""
|
||||
INDEX=`find -L ./ -name "*.amb" | sed 's/.amb//'`
|
||||
|
||||
bwa aln \\
|
||||
$options.args \\
|
||||
-t $task.cpus \\
|
||||
-f ${prefix}.sai \\
|
||||
\$INDEX \\
|
||||
${reads}
|
||||
|
||||
echo \$(bwa 2>&1) | sed 's/^.*Version: //; s/Contact:.*\$//' > ${software}.version.txt
|
||||
"""
|
||||
} else {
|
||||
"""
|
||||
INDEX=`find -L ./ -name "*.amb" | sed 's/.amb//'`
|
||||
|
||||
bwa aln \\
|
||||
$options.args \\
|
||||
-t $task.cpus \\
|
||||
-f ${prefix}.1.sai \\
|
||||
\$INDEX \\
|
||||
${reads[0]}
|
||||
|
||||
bwa aln \\
|
||||
$options.args \\
|
||||
-t $task.cpus \\
|
||||
-f ${prefix}.2.sai \\
|
||||
\$INDEX \\
|
||||
${reads[1]}
|
||||
|
||||
echo \$(bwa 2>&1) | sed 's/^.*Version: //; s/Contact:.*\$//' > ${software}.version.txt
|
||||
"""
|
||||
}
|
||||
}
|
||||
@ -0,0 +1,54 @@
|
||||
name: bwa_aln
|
||||
description: Find SA coordinates of the input reads for bwa short-read mapping
|
||||
keywords:
|
||||
- bwa
|
||||
- aln
|
||||
- short-read
|
||||
- align
|
||||
- reference
|
||||
- fasta
|
||||
- map
|
||||
- fastq
|
||||
tools:
|
||||
- bwa:
|
||||
description: |
|
||||
BWA is a software package for mapping DNA sequences against
|
||||
a large reference genome, such as the human genome.
|
||||
homepage: http://bio-bwa.sourceforge.net/
|
||||
documentation: http://bio-bwa.sourceforge.net/
|
||||
doi: "10.1093/bioinformatics/btp324"
|
||||
licence: ['GPL v3']
|
||||
|
||||
input:
|
||||
- meta:
|
||||
type: map
|
||||
description: |
|
||||
Groovy Map containing sample information
|
||||
e.g. [ id:'test', single_end:false ]
|
||||
- reads:
|
||||
type: file
|
||||
description: |
|
||||
List of input FastQ files of size 1 and 2 for single-end and paired-end data,
|
||||
respectively.
|
||||
- index:
|
||||
type: file
|
||||
description: BWA genome index files
|
||||
pattern: "Directory containing BWA index *.{amb,ann,bwt,pac,sa}"
|
||||
|
||||
output:
|
||||
- meta:
|
||||
type: map
|
||||
description: |
|
||||
Groovy Map containing sample information
|
||||
e.g. [ id:'test', single_end:false ]
|
||||
- version:
|
||||
type: file
|
||||
description: File containing software version
|
||||
pattern: "*.{version.txt}"
|
||||
- sai:
|
||||
type: file
|
||||
description: SA coordinate file
|
||||
pattern: "*.sai"
|
||||
|
||||
authors:
|
||||
- "@jfy133"
|
||||
@ -0,0 +1,33 @@
|
||||
#!/usr/bin/env nextflow
|
||||
|
||||
nextflow.enable.dsl = 2
|
||||
|
||||
include { BWA_INDEX } from '../../../../modules/bwa/index/main.nf' addParams( options: [:] )
|
||||
include { BWA_ALN } from '../../../../modules/bwa/aln/main.nf' addParams( options: [:] )
|
||||
|
||||
//
|
||||
// Test with single-end data
|
||||
//
|
||||
workflow test_bwa_aln_single_end {
|
||||
input = [ [ id:'test', single_end:true ], // meta map
|
||||
[ file(params.test_data['sarscov2']['illumina']['test_1_fastq_gz'], checkIfExists: true) ]
|
||||
]
|
||||
fasta = file(params.test_data['sarscov2']['genome']['genome_fasta'], checkIfExists: true)
|
||||
|
||||
BWA_INDEX ( fasta )
|
||||
BWA_ALN ( input, BWA_INDEX.out.index )
|
||||
}
|
||||
|
||||
//
|
||||
// Test with paired-end data
|
||||
//
|
||||
workflow test_bwa_aln_paired_end {
|
||||
input = [ [ id:'test', single_end:false ], // meta map
|
||||
[ file(params.test_data['sarscov2']['illumina']['test_1_fastq_gz'], checkIfExists: true),
|
||||
file(params.test_data['sarscov2']['illumina']['test_2_fastq_gz'], checkIfExists: true) ]
|
||||
]
|
||||
fasta = file(params.test_data['sarscov2']['genome']['genome_fasta'], checkIfExists: true)
|
||||
|
||||
BWA_INDEX ( fasta )
|
||||
BWA_ALN ( input, BWA_INDEX.out.index )
|
||||
}
|
||||
@ -0,0 +1,39 @@
|
||||
- name: bwa aln single-end
|
||||
command: nextflow run ./tests/modules/bwa/aln -entry test_bwa_aln_single_end -c tests/config/nextflow.config
|
||||
tags:
|
||||
- bwa
|
||||
- bwa/aln
|
||||
files:
|
||||
- path: ./output/bwa/test.sai
|
||||
md5sum: aaaf39b6814c96ca1a5eacc662adf926
|
||||
- path: ./output/index/bwa/genome.bwt
|
||||
md5sum: 0469c30a1e239dd08f68afe66fde99da
|
||||
- path: ./output/index/bwa/genome.amb
|
||||
md5sum: 3a68b8b2287e07dd3f5f95f4344ba76e
|
||||
- path: ./output/index/bwa/genome.ann
|
||||
md5sum: c32e11f6c859f166c7525a9c1d583567
|
||||
- path: ./output/index/bwa/genome.pac
|
||||
md5sum: 983e3d2cd6f36e2546e6d25a0da78d66
|
||||
- path: ./output/index/bwa/genome.sa
|
||||
md5sum: ab3952cabf026b48cd3eb5bccbb636d1
|
||||
|
||||
- name: bwa aln paired-end
|
||||
command: nextflow run ./tests/modules/bwa/aln -entry test_bwa_aln_paired_end -c tests/config/nextflow.config
|
||||
tags:
|
||||
- bwa
|
||||
- bwa/aln
|
||||
files:
|
||||
- path: ./output/bwa/test.1.sai
|
||||
md5sum: aaaf39b6814c96ca1a5eacc662adf926
|
||||
- path: ./output/bwa/test.2.sai
|
||||
md5sum: b4f185d9b4cb256dd5c377070a536124
|
||||
- path: ./output/index/bwa/genome.bwt
|
||||
md5sum: 0469c30a1e239dd08f68afe66fde99da
|
||||
- path: ./output/index/bwa/genome.amb
|
||||
md5sum: 3a68b8b2287e07dd3f5f95f4344ba76e
|
||||
- path: ./output/index/bwa/genome.ann
|
||||
md5sum: c32e11f6c859f166c7525a9c1d583567
|
||||
- path: ./output/index/bwa/genome.pac
|
||||
md5sum: 983e3d2cd6f36e2546e6d25a0da78d66
|
||||
- path: ./output/index/bwa/genome.sa
|
||||
md5sum: ab3952cabf026b48cd3eb5bccbb636d1
|
||||
Loading…
Reference in New Issue