Skip to content
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

Jump to bottom

DSL2: raw library merging + publishing #1082

Merged
merged 4 commits into from
Sep 20, 2024
Merged

DSL2: raw library merging + publishing #1082

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
c18174d
add raw library merging + publishing
scarlhoff Aug 9, 2024
d2983ef
merged input for bedtools, qualimap, sex det
scarlhoff Aug 9, 2024
976c5cf
Apply suggestions from code review
scarlhoff Sep 13, 2024
692e6db
Merge branch 'dev' into dsl2_rawlibrarymerge
scarlhoff Sep 20, 2024
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
73 changes: 58 additions & 15 deletions conf/modules.config
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -730,26 +730,26 @@ process {
// BEDTOOLS_COVERAGE
//
withName: SAMTOOLS_VIEW_GENOME {
tag = { "${meta.reference}|${meta.sample_id}_${meta.library_id}" }
tag = { "${meta.reference}|${meta.sample_id}" }
publishDir = [
enabled: false
]
}

withName: BEDTOOLS_COVERAGE_DEPTH {
tag = { "${meta.reference}|${meta.sample_id}_${meta.library_id}" }
tag = { "${meta.reference}|${meta.sample_id}" }
ext.args = '-mean -nonamecheck'
ext.prefix = { "${meta.sample_id}_${meta.library_id}_${meta.reference}_depth" }
ext.prefix = { "${meta.sample_id}_${meta.reference}_depth" }
publishDir = [
path: { "${params.outdir}/mapstats/bedtools" },
mode: params.publish_dir_mode
]
}

withName: BEDTOOLS_COVERAGE_BREADTH {
tag = { "${meta.reference}|${meta.sample_id}_${meta.library_id}" }
tag = { "${meta.reference}|${meta.sample_id}" }
ext.args = '-nonamecheck'
ext.prefix = { "${meta.sample_id}_${meta.library_id}_${meta.reference}_breadth" }
ext.prefix = { "${meta.sample_id}_${meta.reference}_breadth" }
publishDir = [
path: { "${params.outdir}/mapstats/bedtools" },
mode: params.publish_dir_mode
Expand Down Expand Up @@ -1059,8 +1059,12 @@ process {
ext.args = { "--profiler ${meta.profiler} --output ${meta.profiler}taxpasta_table.tsv" }
}

//
// QUALIMAP
//

withName: 'QUALIMAP_BAMQC_WITHBED|QUALIMAP_BAMQC_NOBED' {
tag = { "${meta.reference}|${meta.sample_id}_${meta.library_id}" }
tag = { "${meta.reference}|${meta.sample_id}" }
publishDir = [
path: { "${params.outdir}/mapstats/qualimap/${meta.reference}/${meta.sample_id}/}" },
mode: params.publish_dir_mode,
Expand Down Expand Up @@ -1107,7 +1111,7 @@ process {
// RUN SEXDETERRMINE
//
withName: SAMTOOLS_DEPTH_SEXDETERRMINE {
tag = { "${meta1.reference}|${meta1.sample_id}_${meta1.library_id}" }
tag = { "${meta1.reference}|${meta1.sample_id}" }
ext.prefix = { "${meta2.id}_samtoolsdepth" }
ext.args = '-aa -q30 -Q30 -H'
publishDir = [
Expand All @@ -1116,7 +1120,7 @@ process {
}

withName: SEXDETERRMINE {
tag = { "${meta.reference}|${meta.sample_id}_${meta.library_id}" }
tag = { "${meta.reference}|${meta.sample_id}" }
ext.prefix = { "${meta.reference}_sexdeterrmine" }
publishDir = [
path: { "${params.outdir}/sex_determination/" },
Expand All @@ -1130,40 +1134,79 @@ process {
// LIBRARY MERGE
//

withName: SAMTOOLS_MERGE_LIBRARIES {
withName: ".*MERGE_LIBRARIES:SAMTOOLS_MERGE_LIBRARIES" {
tag = { "${meta.reference}|${meta.sample_id}" }
ext.prefix = { "${meta.sample_id}_${meta.reference}_unsorted" }
publishDir = [
enabled: false
]
}

withName: ".*MERGE_LIBRARIES:SAMTOOLS_SORT_MERGED_LIBRARIES" {
tag = { "${meta.reference}|${meta.sample_id}" }
ext.prefix = { "${meta.sample_id}_${meta.reference}" }
publishDir = [
path: { "${params.outdir}/final_bams/raw/" },
mode: params.publish_dir_mode,
pattern: '*.bam'
]
}

withName: ".*MERGE_LIBRARIES:SAMTOOLS_INDEX_MERGED_LIBRARIES" {
tag = { "${meta.reference}|${meta.sample_id}" }
ext.args = { params.fasta_largeref ? "-c" : "" }
ext.prefix = { "${meta.sample_id}_${meta.reference}" }
publishDir = [
path: { "${params.outdir}/final_bams/raw/" },
mode: params.publish_dir_mode,
pattern: '*.{bai,csi}'
]
}

withName: ".*MERGE_LIBRARIES:SAMTOOLS_FLAGSTAT_MERGED_LIBRARIES" {
tag = { "${meta.reference}|${meta.sample_id}" }
ext.prefix = { "${meta.sample_id}_${meta.reference}" }
publishDir = [
path: { "${params.outdir}/final_bams/raw/" },
mode: params.publish_dir_mode,
pattern: '*.flagstat'
]
}

withName: ".*MERGE_LIBRARIES:SAMTOOLS_MERGE_LIBRARIES" {
tag = { "${meta.reference}|${meta.sample_id}" }
ext.prefix = { "${meta.sample_id}_${meta.reference}_unsorted" }
publishDir = [
enabled: false
]
}

withName: SAMTOOLS_SORT_MERGED_LIBRARIES {
withName: ".*MERGE_LIBRARIES_GENOTYPING:SAMTOOLS_SORT_MERGED_LIBRARIES" {
tag = { "${meta.reference}|${meta.sample_id}" }
ext.prefix = { "${meta.sample_id}_${meta.reference}" }
publishDir = [
path: { "${params.outdir}/final_bams/" },
path: { "${params.outdir}/final_bams/${params.genotyping_source}/" },
mode: params.publish_dir_mode,
pattern: '*.bam'
]
}

withName: SAMTOOLS_INDEX_MERGED_LIBRARIES {
withName: ".*MERGE_LIBRARIES_GENOTYPING:SAMTOOLS_INDEX_MERGED_LIBRARIES" {
tag = { "${meta.reference}|${meta.sample_id}" }
ext.args = { params.fasta_largeref ? "-c" : "" }
ext.prefix = { "${meta.sample_id}_${meta.reference}" }
publishDir = [
path: { "${params.outdir}/final_bams/" },
path: { "${params.outdir}/final_bams/${params.genotyping_source}/" },
mode: params.publish_dir_mode,
pattern: '*.{bai,csi}'
]
}

withName: SAMTOOLS_FLAGSTAT_MERGED_LIBRARIES {
withName: ".*MERGE_LIBRARIES_GENOTYPING:SAMTOOLS_FLAGSTAT_MERGED_LIBRARIES" {
tag = { "${meta.reference}|${meta.sample_id}" }
ext.prefix = { "${meta.sample_id}_${meta.reference}" }
publishDir = [
path: { "${params.outdir}/final_bams/" },
path: { "${params.outdir}/final_bams/${params.genotyping_source}/" },
mode: params.publish_dir_mode,
pattern: '*.flagstat'
]
Expand Down
66 changes: 36 additions & 30 deletions workflows/eager.nf
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -21,18 +21,19 @@ include { addNewMetaFromAttributes } from '../subworkflows/local/utils_nfcore_ea
//

// TODO rename to active: index_reference, filter_bam etc.
include { REFERENCE_INDEXING } from '../subworkflows/local/reference_indexing'
include { PREPROCESSING } from '../subworkflows/local/preprocessing'
include { MAP } from '../subworkflows/local/map'
include { FILTER_BAM } from '../subworkflows/local/bamfiltering.nf'
include { DEDUPLICATE } from '../subworkflows/local/deduplicate'
include { MANIPULATE_DAMAGE } from '../subworkflows/local/manipulate_damage'
include { METAGENOMICS } from '../subworkflows/local/metagenomics'
include { ESTIMATE_CONTAMINATION } from '../subworkflows/local/estimate_contamination'
include { CALCULATE_DAMAGE } from '../subworkflows/local/calculate_damage'
include { RUN_SEXDETERRMINE } from '../subworkflows/local/run_sex_determination'
include { MERGE_LIBRARIES } from '../subworkflows/local/merge_libraries'
include { GENOTYPE } from '../subworkflows/local/genotype'
include { REFERENCE_INDEXING } from '../subworkflows/local/reference_indexing'
include { PREPROCESSING } from '../subworkflows/local/preprocessing'
include { MAP } from '../subworkflows/local/map'
include { FILTER_BAM } from '../subworkflows/local/bamfiltering.nf'
include { DEDUPLICATE } from '../subworkflows/local/deduplicate'
include { MANIPULATE_DAMAGE } from '../subworkflows/local/manipulate_damage'
include { METAGENOMICS_COMPLEXITYFILTER } from '../subworkflows/local/metagenomics_complexityfilter'
include { ESTIMATE_CONTAMINATION } from '../subworkflows/local/estimate_contamination'
include { CALCULATE_DAMAGE } from '../subworkflows/local/calculate_damage'
include { RUN_SEXDETERRMINE } from '../subworkflows/local/run_sex_determination'
include { MERGE_LIBRARIES } from '../subworkflows/local/merge_libraries'
include { MERGE_LIBRARIES as MERGE_LIBRARIES_GENOTYPING } from '../subworkflows/local/merge_libraries'
include { GENOTYPE } from '../subworkflows/local/genotype'

/*
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Expand Down Expand Up @@ -264,6 +265,15 @@ workflow EAGER {
ch_dedupped_flagstat = Channel.empty()
}

//
// SUBWORKFLOW: Merge libraries per sample
//

MERGE_LIBRARIES ( ch_dedupped_bams )
ch_versions = ch_versions.mix( MERGE_LIBRARIES.out.versions )
ch_merged_dedup_bams = MERGE_LIBRARIES.out.bam_bai
ch_multiqc_files = ch_multiqc_files.mix( MERGE_LIBRARIES.out.mqc.collect{it[1]}.ifEmpty([]) )

//
// MODULE QUALIMAP
//
Expand All @@ -273,7 +283,7 @@ workflow EAGER {
.map{
addNewMetaFromAttributes( it, "id" , "reference" , false )
}
ch_qualimap_input = ch_dedupped_bams
ch_qualimap_input = ch_merged_dedup_bams
.map {
meta, bam, bai ->
[ meta, bam ]
Expand Down Expand Up @@ -458,7 +468,7 @@ workflow EAGER {
addNewMetaFromAttributes( it, "id" , "reference" , false )
}

ch_bedtools_prep = ch_dedupped_bams
ch_bedtools_prep = ch_merged_dedup_bams
.map {
addNewMetaFromAttributes( it, "reference" , "reference" , false )
}
Expand Down Expand Up @@ -518,7 +528,7 @@ workflow EAGER {
//

if ( params.run_sexdeterrmine ) {
ch_sexdeterrmine_input = ch_dedupped_bams
ch_sexdeterrmine_input = ch_merged_dedup_bams

RUN_SEXDETERRMINE(ch_sexdeterrmine_input, REFERENCE_INDEXING.out.sexdeterrmine_bed )
ch_versions = ch_versions.mix( RUN_SEXDETERRMINE.out.versions )
Expand All @@ -539,27 +549,23 @@ workflow EAGER {

//
// SUBWORKFLOW: aDNA Damage Manipulation
//

if ( params.run_mapdamage_rescaling || params.run_pmd_filtering || params.run_trim_bam ) {
MANIPULATE_DAMAGE( ch_dedupped_bams, ch_fasta_for_deduplication.fasta, REFERENCE_INDEXING.out.pmd_masking )
ch_multiqc_files = ch_multiqc_files.mix( MANIPULATE_DAMAGE.out.flagstat.collect{it[1]}.ifEmpty([]) )
ch_versions = ch_versions.mix( MANIPULATE_DAMAGE.out.versions )
ch_bams_for_library_merge = params.genotyping_source == 'rescaled' ? MANIPULATE_DAMAGE.out.rescaled : params.genotyping_source == 'pmd' ? MANIPULATE_DAMAGE.out.filtered : params.genotyping_source == 'trimmed' ? MANIPULATE_DAMAGE.out.trimmed : ch_dedupped_bams
ch_multiqc_files = ch_multiqc_files.mix( MANIPULATE_DAMAGE.out.flagstat.collect{it[1]}.ifEmpty([]) )
ch_versions = ch_versions.mix( MANIPULATE_DAMAGE.out.versions )
ch_bams_for_library_merge = params.genotyping_source == 'rescaled' ? MANIPULATE_DAMAGE.out.rescaled : params.genotyping_source == 'pmd' ? MANIPULATE_DAMAGE.out.filtered : params.genotyping_source == 'trimmed' ? MANIPULATE_DAMAGE.out.trimmed : ch_merged_dedup_bams

// SUBWORKFLOW: merge libraries for genotyping
MERGE_LIBRARIES_GENOTYPING ( ch_bams_for_library_merge )
ch_versions = ch_versions.mix( MERGE_LIBRARIES_GENOTYPING.out.versions )
ch_bams_for_genotyping = MERGE_LIBRARIES_GENOTYPING.out.bam_bai
ch_multiqc_files = ch_multiqc_files.mix( MERGE_LIBRARIES_GENOTYPING.out.mqc.collect{it[1]}.ifEmpty([]) )
} else {
ch_bams_for_library_merge = ch_dedupped_bams
ch_bams_for_genotyping = ch_merged_dedup_bams
}

//
// SUBWORKFLOW: MERGE LIBRARIES
//

// The bams being merged are always the ones specified by params.genotyping_source,
// unless the user skipped damage manipulation, in which case it is the DEDUPLICATION output.
MERGE_LIBRARIES ( ch_bams_for_library_merge )
ch_versions = ch_versions.mix( MERGE_LIBRARIES.out.versions )
ch_bams_for_genotyping = MERGE_LIBRARIES.out.bam_bai
ch_multiqc_files = ch_multiqc_files.mix( MERGE_LIBRARIES.out.mqc.collect{it[1]}.ifEmpty([]) ) // Not sure if this is needed, or if it needs to be moved to line 564?

//
// SUBWORKFLOW: Genotyping
//
Expand Down