Skip to content

Make sure HISAT2 index is not built when provided via parameter #196

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
nictru merged 2 commits into from
Apr 22, 2025
Merged

Make sure HISAT2 index is not built when provided via parameter #196

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
71ec63e
Remove all when directives from modules.config
nictru Apr 22, 2025
7cca75f
Make sure hisat2 and star are executed under the correct conditions
nictru Apr 22, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
17 changes: 5 additions & 12 deletions conf/modules.config
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -110,7 +110,6 @@ process {
}

withName: BOWTIE_BUILD {
ext.when = { !params.bowtie && params.tools.split(',').contains('mapsplice') }
publishDir = [
path: { "${params.outdir}/1_references/index" },
mode: params.publish_dir_mode,
Expand All @@ -119,7 +118,6 @@ process {
}

withName: BOWTIE2_BUILD {
ext.when = { !params.bowtie2 && params.tools.split(',').contains('find_circ') }
publishDir = [
path: { "${params.outdir}/1_references/index" },
mode: params.publish_dir_mode,
Expand All @@ -128,7 +126,6 @@ process {
}

withName: BWA_INDEX {
ext.when = { !params.bwa && params.tools.split(',').contains('ciriquant') }
publishDir = [
path: { "${params.outdir}/1_references/index" },
mode: params.publish_dir_mode,
Expand All @@ -137,7 +134,6 @@ process {
}

withName: HISAT2_EXTRACTSPLICESITES {
ext.when = { params.tools.split(',').contains('ciriquant') }
publishDir = [
path: { "${params.outdir}/1_references/index/hisat2" },
mode: params.publish_dir_mode,
Expand All @@ -146,7 +142,6 @@ process {
}

withName: HISAT2_BUILD {
ext.when = { params.tools.split(',').contains('ciriquant') }
publishDir = [
path: { "${params.outdir}/1_references/index" },
mode: params.publish_dir_mode,
Expand All @@ -155,7 +150,6 @@ process {
}

withName: STAR_GENOMEGENERATE {
ext.when = { !params.star && (params.tools.split(',').contains('circexplorer2') || params.tools.split(',').contains('dcc') || params.tools.split(',').contains('circrna_finder')) }
ext.args = [
"",
params.sjdboverhang ? "--sjdbOverhang ${params.sjdboverhang}" : '',
Expand Down Expand Up @@ -243,7 +237,6 @@ process {
}

withName: '.*STAR2PASS:PASS_1' {
ext.when = { params.tools.split(',').contains('circexplorer2') || params.tools.split(',').contains('circrna_finder') }
ext.args = [
"",
"--chimOutType Junctions WithinBAM",
Expand Down Expand Up @@ -579,14 +572,14 @@ process {
publishDir = [
path: { "${params.outdir}/3_bsj_detection/samples/${meta.id}" },
mode: params.publish_dir_mode,
saveAs: { filename -> (filename.equals('versions.yml') || filename.endsWith('_mqc.json')) ? null : filename },
saveAs: { filename -> filename.equals('versions.yml') || filename.endsWith('_mqc.json') ? null : filename },
]
}

withName: COMBINEBEDS_READS {
ext.consider_strand = params.consider_strand
ext.prefix = { "${meta.id}_reads" }
publishDir = [
ext.prefix = { "${meta.id}_reads" }
publishDir = [
path: { "${params.outdir}/3_bsj_detection/samples/${meta.id}/reads" },
mode: params.publish_dir_mode,
saveAs: { filename -> filename.equals('versions.yml') ? null : filename },
Expand All @@ -606,7 +599,7 @@ process {
publishDir = [
path: { "${params.outdir}/3_bsj_detection/combined" },
mode: params.publish_dir_mode,
saveAs: { filename -> (filename.equals('versions.yml') || filename.endsWith('_mqc.json')) ? null : filename },
saveAs: { filename -> filename.equals('versions.yml') || filename.endsWith('_mqc.json') ? null : filename },
]
}

Expand Down Expand Up @@ -1071,7 +1064,7 @@ process {
}

withName: MULTIQC {
ext.args = { params.multiqc_title ? "--title \"${params.multiqc_title}\"" : '' }
ext.args = { params.multiqc_title ? "--title \"${params.multiqc_title}\"" : '' }
publishDir = [
path: { "${params.outdir}/7_multiqc" },
mode: params.publish_dir_mode,
Expand Down
10 changes: 7 additions & 3 deletions subworkflows/local/bsj_detection.nf
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -47,17 +47,21 @@ workflow BSJ_DETECTION {
fasta = ch_fasta.map{_meta, fasta -> fasta}
gtf = ch_gtf.map{_meta, gtf -> gtf}

tools_selected = params.tools.split(',').collect{it.trim().toLowerCase()}

// STAR 2-PASS-MODE
star_ignore_sjdbgtf = true
seq_center = params.seq_center ?: ''
seq_platform = ''
STAR2PASS( reads, star_index, ch_gtf, bsj_reads, star_ignore_sjdbgtf, seq_center, seq_platform )
ch_versions = ch_versions.mix(STAR2PASS.out.versions)

if (tools_selected.intersect(['circexplorer2', 'circrna_finder', 'dcc', 'mapsplice']).size() > 0) {
STAR2PASS( reads, star_index, ch_gtf, bsj_reads, star_ignore_sjdbgtf, seq_center, seq_platform )
ch_versions = ch_versions.mix(STAR2PASS.out.versions)
}

//
// DISCOVERY TOOLS:
//
tools_selected = params.tools.split(',').collect{it.trim().toLowerCase()}

if (tools_selected.size() == 0) {
error 'No tools selected for circRNA discovery.'
Expand Down
52 changes: 30 additions & 22 deletions subworkflows/local/prepare_genome.nf
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -12,19 +12,18 @@ include { UCSC_GTFTOGENEPRED } from '../../modules/nf-core/ucsc/gtf
include { GAWK as CIRCEXPLORER2_REFERENCE } from '../../modules/nf-core/gawk'

workflow PREPARE_GENOME {

take:
ch_fasta
ch_gtf

main:
ch_versions = Channel.empty()

detection_tools = params.tools.split(',').collect{it.trim().toLowerCase()}
detection_tools = params.tools.split(',').collect { it.trim().toLowerCase() }

// MapSplice cannot deal with extra field in the fasta headers
// this removes all additional fields in the headers of the input fasta file
if( detection_tools.contains('mapsplice') ) {
if (detection_tools.contains('mapsplice')) {
CLEAN_FASTA(ch_fasta, [], false)
ch_fasta = CLEAN_FASTA.out.output

Expand All @@ -41,45 +40,54 @@ workflow PREPARE_GENOME {
SEQKIT_SPLIT(ch_fasta)
ch_versions = ch_versions.mix(SEQKIT_SPLIT.out.versions)

BOWTIE_BUILD(ch_fasta)
ch_versions = ch_versions.mix(BOWTIE_BUILD.out.versions)
if (!params.bowtie && detection_tools.contains('mapsplice')) {
BOWTIE_BUILD(ch_fasta)
ch_versions = ch_versions.mix(BOWTIE_BUILD.out.versions)
}

BOWTIE2_BUILD(ch_fasta)
ch_versions = ch_versions.mix(BOWTIE2_BUILD.out.versions)
if (!params.bowtie2 && detection_tools.contains('find_circ')) {
BOWTIE2_BUILD(ch_fasta)
ch_versions = ch_versions.mix(BOWTIE2_BUILD.out.versions)
}

BWA_INDEX (ch_fasta)
ch_versions = ch_versions.mix(BWA_INDEX.out.versions)
if (!params.bwa && detection_tools.contains('ciriquant')) {
BWA_INDEX(ch_fasta)
ch_versions = ch_versions.mix(BWA_INDEX.out.versions)
}

HISAT2_EXTRACTSPLICESITES(ch_gtf)
ch_versions = ch_versions.mix(HISAT2_EXTRACTSPLICESITES.out.versions)
if (!params.hisat2 && detection_tools.contains('ciriquant')) {
HISAT2_EXTRACTSPLICESITES(ch_gtf)
ch_versions = ch_versions.mix(HISAT2_EXTRACTSPLICESITES.out.versions)

HISAT2_BUILD(ch_fasta, ch_gtf, HISAT2_EXTRACTSPLICESITES.out.txt)
ch_versions = ch_versions.mix(HISAT2_BUILD.out.versions)
HISAT2_BUILD(ch_fasta, ch_gtf, HISAT2_EXTRACTSPLICESITES.out.txt)
ch_versions = ch_versions.mix(HISAT2_BUILD.out.versions)
}

STAR_GENOMEGENERATE(ch_fasta, ch_gtf)
ch_versions = ch_versions.mix(STAR_GENOMEGENERATE.out.versions)
if (!params.star && detection_tools.intersect(['circexplorer2', 'circrna_finder', 'dcc', 'mapsplice']).size() > 0) {
STAR_GENOMEGENERATE(ch_fasta, ch_gtf)
ch_versions = ch_versions.mix(STAR_GENOMEGENERATE.out.versions)
}

SAMTOOLS_FAIDX(ch_fasta, [[], []])
ch_versions = ch_versions.mix(SAMTOOLS_FAIDX.out.versions)

ch_circexplorer2_reference = Channel.empty()
if (detection_tools.intersect(['circexplorer2', 'mapsplice']).size() > 0) {
CIRCEXPLORER2_REFERENCE(UCSC_GTFTOGENEPRED.out.genepred, [], false)
ch_circexplorer2_reference = CIRCEXPLORER2_REFERENCE.out.output.map{ _meta, file -> file }.collect()
ch_circexplorer2_reference = CIRCEXPLORER2_REFERENCE.out.output.map { _meta, file -> file }.collect()
ch_versions = ch_versions.mix(CIRCEXPLORER2_REFERENCE.out.versions)
}

emit:
gtf = ch_gtf
faidx = SAMTOOLS_FAIDX.out.fai
bowtie = params.bowtie ?: BOWTIE_BUILD.out.index
bowtie2 = params.bowtie2 ? Channel.value([[id: "bowtie2"], file(params.bowtie2, checkIfExists: true)]) : BOWTIE2_BUILD.out.index.collect()
bwa = params.bwa ? Channel.value([[id: "bwa"], file(params.bwa, checkIfExists: true)]) : BWA_INDEX.out.index.collect()
hisat2 = params.hisat2 ? Channel.value([[id: "hisat2"], file(params.hisat2, checkIfExists: true)]) : HISAT2_BUILD.out.index.collect()
star = params.star ? Channel.value([[id: "star"], file(params.star, checkIfExists: true)]) : STAR_GENOMEGENERATE.out.index.collect()
bowtie = params.bowtie ?: BOWTIE_BUILD.out.index
bowtie2 = params.bowtie2 ? Channel.value([[id: "bowtie2"], file(params.bowtie2, checkIfExists: true)]) : BOWTIE2_BUILD.out.index.collect()
bwa = params.bwa ? Channel.value([[id: "bwa"], file(params.bwa, checkIfExists: true)]) : BWA_INDEX.out.index.collect()
hisat2 = params.hisat2 ? Channel.value([[id: "hisat2"], file(params.hisat2, checkIfExists: true)]) : HISAT2_BUILD.out.index.collect()
star = params.star ? Channel.value([[id: "star"], file(params.star, checkIfExists: true)]) : STAR_GENOMEGENERATE.out.index.collect()
circexplorer2 = ch_circexplorer2_reference
chromosomes = SEQKIT_SPLIT.out.split
splice_sites = HISAT2_EXTRACTSPLICESITES.out.txt.collect()

versions = ch_versions
}
Loading