Skip to content

Commit

Permalink
port parabricks/deepvariant to nf-test (#6995)
Browse files Browse the repository at this point in the history 
* port to nf-test

* remove

* update test yaml

* swap to nextflow config

* finalize

* revert changes to nf-test config

* revert changes to nf-test config

* add to conda_skip

* add tag gpu

* Update modules/nf-core/parabricks/deepvariant/tests/nextflow.config

* Update modules/nf-core/parabricks/deepvariant/main.nf

* fix tests and add myself as maintainer to other modules

* fix meta

* add isssue to comment

* remove conda skip

* update snap

---------

Co-authored-by: Simon Pearce <24893913+SPPearce@users.noreply.github.com>
  • Loading branch information
@famosab @SPPearce
famosab and SPPearce authored Jan 16, 2025
1 parent 070b1ad commit f549c75
Show file tree
Hide file tree
Showing 16 changed files with 829 additions and 150 deletions.
1 change: 1 addition & 0 deletions modules/nf-core/parabricks/applybqsr/meta.yml
View file
Original file line number Diff line number Diff line change
Expand Up @@ -88,3 +88,4 @@ authors:
- "@bsiranosian"
maintainers:
- "@bsiranosian"
- "@famosab"
2 changes: 2 additions & 0 deletions modules/nf-core/parabricks/dbsnp/meta.yml
View file
Original file line number Diff line number Diff line change
Expand Up @@ -50,3 +50,5 @@ output:
pattern: "versions.yml"
authors:
- "@Furentsu"
maintainers:
- "@famosab"
22 changes: 11 additions & 11 deletions modules/nf-core/parabricks/deepvariant/main.nf
View file
Original file line number Diff line number Diff line change
@@ -1,40 +1,41 @@
process PARABRICKS_DEEPVARIANT {
tag "$meta.id"
label 'process_high'
label 'process_gpu'
stageInMode 'copy' // needed by the module to work properly can be removed when fixed upstream - Issue #7226

container "nvcr.io/nvidia/clara/clara-parabricks:4.3.0-1"
container "nvcr.io/nvidia/clara/clara-parabricks:4.4.0-1"

input:
tuple val(meta), path(input), path(input_index), path(interval_file)
tuple val(ref_meta), path(fasta)

output:
tuple val(meta), path("*.vcf"), emit: vcf
path "versions.yml", emit: versions
tuple val(meta), path("*.vcf"), optional: true, emit: vcf
tuple val(meta), path("*.g.vcf.gz"), optional: true, emit: gvcf
path "versions.yml", emit: versions

when:
task.ext.when == null || task.ext.when

script:

// Exit if running this module with -profile conda / -profile mamba
if (workflow.profile.tokenize(',').intersect(['conda', 'mamba']).size() >= 1) {
exit 1, "Parabricks module does not support Conda. Please use Docker / Singularity / Podman instead."
}

def args = task.ext.args ?: ''
def prefix = task.ext.prefix ?: "${meta.id}"
def output_file = args =~ "gvcf" ? "${prefix}.g.vcf" : "${prefix}.vcf"
def output_file = ("--gvcf" =~ task.ext.args)? "${prefix}.g.vcf.gz" : "${prefix}.vcf"
def interval_file_command = interval_file ? interval_file.collect{"--interval-file $it"}.join(' ') : ""
def num_gpus = task.accelerator ? "--num-gpus $task.accelerator.request" : ''
"""
pbrun \\
deepvariant \\
--ref $fasta \\
--in-bam $input \\
--out-variants $output_file \\
$interval_file_command \\
--num-gpus $task.accelerator.request \\
$num_gpus \\
$args
cat <<-END_VERSIONS > versions.yml
Expand All @@ -44,11 +45,10 @@ process PARABRICKS_DEEPVARIANT {
"""

stub:
def args = task.ext.args ?: ''
def prefix = task.ext.prefix ?: "${meta.id}"
def output_file = args =~ "gvcf" ? "${prefix}.g.vcf" : "${prefix}.vcf"
def output_cmd = ("--gvcf" =~ task.ext.args)? "echo '' | gzip > ${prefix}.g.vcf.gz" : "touch ${prefix}.vcf"
"""
touch $output_file
$output_cmd
cat <<-END_VERSIONS > versions.yml
"${task.process}":
Expand Down
14 changes: 13 additions & 1 deletion modules/nf-core/parabricks/deepvariant/meta.yml
View file
Original file line number Diff line number Diff line change
Expand Up @@ -53,12 +53,24 @@ output:
e.g. [ id:'test' ]
- "*.vcf":
type: file
description: Variant file.
description: vcf file created with deepvariant (does not support .gz for normal vcf), optional
pattern: "*.vcf"
- gvcf:
- meta:
type: map
description: |
Groovy Map containing sample information.
e.g. [ id:'test' ]
- "*.g.vcf.gz":
type: file
description: bgzipped gvcf created with deepvariant, optional
pattern: "*.g.vcf.gz"
- versions:
- versions.yml:
type: file
description: File containing software versions.
pattern: "versions.yml"
authors:
- "@bsiranosian"
maintainers:
- "@famosab"
223 changes: 223 additions & 0 deletions modules/nf-core/parabricks/deepvariant/tests/main.nf.test
View file
Original file line number Diff line number Diff line change
@@ -0,0 +1,223 @@
nextflow_process {

name "Test Process PARABRICKS_DEEPVARIANT"
script "../main.nf"
process "PARABRICKS_DEEPVARIANT"

tag "modules"
tag "modules_nfcore"
tag "parabricks"
tag "parabricks/deepvariant"
tag "gpu"

test("human - bam") {

when {
process {
"""
input[0] = [
[ id:'test'],
file(params.modules_testdata_base_path + 'genomics/homo_sapiens/illumina/bam/test2.paired_end.recalibrated.sorted.bam', checkIfExists: true),
[], // bam.bai not needed unless using intervals
[]
]
input[1] = [
[ id:'test'],
file(params.modules_testdata_base_path + 'genomics/homo_sapiens/genome/chr21/sequence/genome.fasta', checkIfExists: true)
]
"""
}
}

then {
assertAll(
{ assert process.success },
{ assert snapshot(
path(process.out.vcf[0][1]).vcf.getVariantsAsStrings(100),
process.out.versions,
path(process.out.versions[0]).yaml
).match() }
)
}

}

test("human - bam - intervals") {

when {
process {
"""
input[0] = [
[ id:'test'],
file(params.modules_testdata_base_path + 'genomics/homo_sapiens/illumina/bam/test2.paired_end.recalibrated.sorted.bam', checkIfExists: true),
file(params.modules_testdata_base_path + 'genomics/homo_sapiens/illumina/bam/test2.paired_end.recalibrated.sorted.bam.bai', checkIfExists: true),
file(params.modules_testdata_base_path + 'genomics/homo_sapiens/genome/chr21/sequence/multi_intervals.bed', checkIfExists: true)
]
input[1] = [
[ id:'ref'],
file(params.modules_testdata_base_path + 'genomics/homo_sapiens/genome/chr21/sequence/genome.fasta', checkIfExists: true)
]
"""
}
}

then {
assertAll(
{ assert process.success },
{ assert snapshot(
path(process.out.vcf[0][1]).vcf.getVariantsAsStrings(100),
process.out.versions,
path(process.out.versions[0]).yaml
).match() }
)
}

}

test("human - bam - gvcf") {

config './nextflow.config'

when {
params {
module_args = '--gvcf'
}
process {
"""
input[0] = [
[ id:'test'],
file(params.modules_testdata_base_path + 'genomics/homo_sapiens/illumina/bam/test2.paired_end.recalibrated.sorted.bam', checkIfExists: true),
[], // bam.bai not needed unless using intervals
[]
]
input[1] = [
[ id:'ref'],
file(params.modules_testdata_base_path + 'genomics/homo_sapiens/genome/chr21/sequence/genome.fasta', checkIfExists: true)
]
"""
}
}

then {
assertAll(
{ assert process.success },
{ assert snapshot(
path(process.out.gvcf[0][1]).vcf.getVariantsAsStrings(100),
process.out.versions,
path(process.out.versions[0]).yaml
).match()
}
)
}

}

test("human - bam - intervals - gvcf") {

config './nextflow.config'

when {
params {
module_args = '--gvcf'
}
process {
"""
input[0] = [
[ id:'test'],
file(params.modules_testdata_base_path + 'genomics/homo_sapiens/illumina/bam/test2.paired_end.recalibrated.sorted.bam', checkIfExists: true),
file(params.modules_testdata_base_path + 'genomics/homo_sapiens/illumina/bam/test2.paired_end.recalibrated.sorted.bam.bai', checkIfExists: true),
file(params.modules_testdata_base_path + 'genomics/homo_sapiens/genome/chr21/sequence/multi_intervals.bed', checkIfExists: true)
]
input[1] = [
[ id:'ref'],
file(params.modules_testdata_base_path + 'genomics/homo_sapiens/genome/chr21/sequence/genome.fasta', checkIfExists: true)
]
"""
}
}

then {
assertAll(
{ assert process.success },
{ assert snapshot(
path(process.out.gvcf[0][1]).vcf.getVariantsAsStrings(100),
process.out.versions,
path(process.out.versions[0]).yaml
).match()
}
)
}

}

test("human - bam - stub") {

options "-stub"

when {
process {
"""
input[0] = [
[ id:'test'],
file(params.modules_testdata_base_path + 'genomics/homo_sapiens/illumina/bam/test2.paired_end.recalibrated.sorted.bam', checkIfExists: true),
[], // bam.bai not needed unless using intervals
[]
]
input[1] = [
[ id:'test'],
file(params.modules_testdata_base_path + 'genomics/homo_sapiens/genome/chr21/sequence/genome.fasta', checkIfExists: true)
]
"""
}
}

then {
assertAll(
{ assert process.success },
{ assert snapshot(
process.out,
path(process.out.versions[0]).yaml
).match() }
)
}

}

test("human - bam - intervals - gvcf - stub") {

config './nextflow.config'
options "-stub"

when {
params {
module_args = '--gvcf'
}
process {
"""
input[0] = [
[ id:'test'],
file(params.modules_testdata_base_path + 'genomics/homo_sapiens/illumina/bam/test2.paired_end.recalibrated.sorted.bam', checkIfExists: true),
file(params.modules_testdata_base_path + 'genomics/homo_sapiens/illumina/bam/test2.paired_end.recalibrated.sorted.bam.bai', checkIfExists: true),
file(params.modules_testdata_base_path + 'genomics/homo_sapiens/genome/chr21/sequence/multi_intervals.bed', checkIfExists: true)
]
input[1] = [
[ id:'test'],
file(params.modules_testdata_base_path + 'genomics/homo_sapiens/genome/chr21/sequence/genome.fasta', checkIfExists: true)
]
"""
}
}

then {
assertAll(
{ assert process.success },
{ assert snapshot(
process.out,
path(process.out.versions[0]).yaml
).match() }
)
}

}

}
Loading

0 comments on commit f549c75

Please sign in to comment.