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

Small updates to JointVcfFiltering WDL #8027

Merged
merged 3 commits into from
Sep 22, 2022
Merged

Small updates to JointVcfFiltering WDL #8027

Changes from 2 commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
6305845
Small changes to JointVCFFiltering WDL
meganshand Sep 20, 2022
718cda4
making default for use_allele_specific_annotations
meganshand Sep 21, 2022
5b61044
addressing comments
meganshand Sep 22, 2022
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Jump to
Jump to file
Failed to load files.
Diff view
Diff view
54 changes: 36 additions & 18 deletions scripts/vcf_site_level_filtering_wdl/JointVcfFiltering.wdl
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -27,6 +27,8 @@ workflow JointVcfFiltering {
String indel_annotations
File? gatk_override

Boolean use_allele_specific_annotations = false
Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I gave this a default, but I can also see an argument for having it be a required input since it's somewhat tied to the input annotations (which are currently required). Would love a second opinion.

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Good point. I think it makes sense to make it required.

More generally, for task-level required/optional arguments, I like the style we use in the CNV WDLs. That is, all arguments are exposed, and optional arguments have their defaults specified in the command line rather than in the parameter block (see e.g.

gatk/scripts/cnv_wdl/cnv_common_tasks.wdl

Line 34 in ef50c08

gatk --java-options "-Xmx~{command_mem_mb}m" PreprocessIntervals \
). This makes it easier to glance at the command line and see the defaults.

Not sure if this is the way automatic WDL generation works (haven't tried it myself), but I think it makes sense. And ideally we'd be good about using that functionality that to keep the defaults in sync between the WDL and the jar. Baby steps!

Don't worry about all that for this PR---I'll tackle it in my push. Thanks!

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Ah makes sense, it does seem cleaner to keep the defaults in the command line.


String snp_resource_args = "--resource:hapmap,training=true,calibration=true gs://gcp-public-data--broad-references/hg38/v0/hapmap_3.3.hg38.vcf.gz --resource:omni,training=true,calibration=true gs://gcp-public-data--broad-references/hg38/v0/1000G_omni2.5.hg38.vcf.gz --resource:1000G,training=true,calibration=false gs://gcp-public-data--broad-references/hg38/v0/1000G_phase1.snps.high_confidence.hg38.vcf.gz"
String indel_resource_args = "--resource:mills,training=true,calibration=true gs://gcp-public-data--broad-references/hg38/v0/Mills_and_1000G_gold_standard.indels.hg38.vcf.gz"
}
Expand All @@ -46,6 +48,7 @@ workflow JointVcfFiltering {
resource_args = snp_resource_args,
basename = basename,
interval_list = extract_interval_list,
use_allele_specific_annotations = use_allele_specific_annotations,
gatk_override = gatk_override,
gatk_docker = gatk_docker
}
Expand All @@ -59,6 +62,7 @@ workflow JointVcfFiltering {
resource_args = indel_resource_args,
basename = basename,
interval_list = extract_interval_list,
use_allele_specific_annotations = use_allele_specific_annotations,
gatk_override = gatk_override,
gatk_docker = gatk_docker
}
Expand Down Expand Up @@ -102,6 +106,7 @@ workflow JointVcfFiltering {
interval_list = score_interval_list,
model_files = TrainVariantAnnotationModelSNPs.outputs,
resource_args = snp_resource_args,
use_allele_specific_annotations = use_allele_specific_annotations,
gatk_override = gatk_override,
gatk_docker = gatk_docker
}
Expand All @@ -120,14 +125,16 @@ workflow JointVcfFiltering {
interval_list = score_interval_list,
model_files = TrainVariantAnnotationModelINDELs.outputs,
resource_args = indel_resource_args,
use_allele_specific_annotations = use_allele_specific_annotations,
gatk_override = gatk_override,
gatk_docker = gatk_docker
}

}

output {
Array[File] variant_filtered_vcf = ScoreVariantAnnotationsINDELs.output_vcf
Array[File] variant_filtered_vcf_index = ScoreVariantAnnotationsINDELs.output_vcf_index
Array[File] variant_scored_vcf = ScoreVariantAnnotationsINDELs.output_vcf
Array[File] variant_scored_vcf_index = ScoreVariantAnnotationsINDELs.output_vcf_index
}

}
Expand All @@ -143,6 +150,7 @@ task ExtractVariantAnnotations {
String annotations
String resource_args
File? interval_list
Boolean use_allele_specific_annotations

Int memory_mb = 14000
Int command_mem = memory_mb - 1000
Expand All @@ -157,6 +165,7 @@ task ExtractVariantAnnotations {
-V ~{input_vcf} \
-O ~{basename}.~{mode} \
~{annotations} \
~{if use_allele_specific_annotations then "--use-allele-specific-annotations" else ""} \
~{"-L " + interval_list} \
--mode ~{mode} \
~{resource_args}
Expand Down Expand Up @@ -232,35 +241,44 @@ task ScoreVariantAnnotations {
File extracted_training_vcf_index
File? interval_list
Array[File] model_files
Boolean use_allele_specific_annotations

Int memory_mb = 16000
Int command_mem = memory_mb - 1000
}
Int disk_size = ceil(size(vcf, "GB") *2 + 50)

command {
zgrep -v '#' ~{vcf} > empty.txt
set -e

ln -s ~{sep=" . && ln -s " model_files} .
if [ -s empty.txt ]; then
ln -s ~{sep=" . && ln -s " model_files} .

export GATK_LOCAL_JAR=~{default="/root/gatk.jar" gatk_override}
export GATK_LOCAL_JAR=~{default="/root/gatk.jar" gatk_override}

gatk --java-options "-Xmx~{command_mem}m" \
ScoreVariantAnnotations \
~{"-L " + interval_list} \
-V ~{vcf} \
-O ~{basename}.~{mode} \
--model-backend ~{model_backend} \
~{"--python-script " + python_script} \
--model-prefix ~{basename} \
~{annotations} \
--mode ~{mode} \
--resource:extracted,extracted=true ~{extracted_training_vcf} \
~{resource_args}
gatk --java-options "-Xmx~{command_mem}m" \
ScoreVariantAnnotations \
~{"-L " + interval_list} \
-V ~{vcf} \
-O ~{basename}.~{mode} \
--model-backend ~{model_backend} \
~{"--python-script " + python_script} \
--model-prefix ~{basename} \
~{annotations} \
~{if use_allele_specific_annotations then "--use-allele-specific-annotations" else ""} \
-mode ~{mode} \
--resource:extracted,extracted=true ~{extracted_training_vcf} \
~{resource_args}
else
echo "Input VCF was empty so we'll return the same VCF that was input"
ln -s ~{vcf} ~{basename}.~{mode}.vcf.gz
ln -s ~{vcf_index} ~{basename}.~{mode}.vcf.gz.tbi
fi
}
output {
File scores = "~{basename}.~{mode}.scores.hdf5"
File annots = "~{basename}.~{mode}.annot.hdf5"
File? scores = "~{basename}.~{mode}.scores.hdf5"
meganshand marked this conversation as resolved.
Show resolved Hide resolved
File? annots = "~{basename}.~{mode}.annot.hdf5"
File output_vcf = "~{basename}.~{mode}.vcf.gz"
File output_vcf_index = "~{basename}.~{mode}.vcf.gz.tbi"
}
Expand Down