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

PIPE-52-gatk-reference-tarball #146

Merged
merged 3 commits into from
Mar 3, 2022
Merged

PIPE-52-gatk-reference-tarball #146

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
99c0b8f
PIPE-52-gatk-reference-tarball
paul-sud Mar 1, 2022
fa2b4ca
fix bundle unpacking
paul-sud Mar 2, 2022
0431949
bump versions
paul-sud Mar 3, 2022
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
60 changes: 13 additions & 47 deletions genophase.wdl
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4,38 +4,29 @@ import "./hic.wdl"

workflow genophase {
meta {
version: "1.11.2"
caper_docker: "encodedcc/hic-pipeline:1.11.2"
caper_singularity: "docker://encodedcc/hic-pipeline:1.11.2"
version: "1.11.3"
caper_docker: "encodedcc/hic-pipeline:1.11.3"
caper_singularity: "docker://encodedcc/hic-pipeline:1.11.3"
croo_out_def: "https://raw.githubusercontent.com/ENCODE-DCC/hic-pipeline/dev/croo_out_def.json"
}

input {
File reference_fasta
Array[File] bams
# From GATK bundle
# .tar.gz archive containing Ommi, Mills, Hapmap, and 1000G VCFs + indexes
# https://gatk.broadinstitute.org/hc/en-us/articles/360035890811-Resource-bundle
# https://console.cloud.google.com/storage/browser/genomics-public-data/resources/broad/hg38/v0/
File dbsnp_vcf
File dbsnp_vcf_index
File hapmap_vcf_index
File hapmap_vcf
File mills_vcf
File mills_vcf_index
File omni_vcf
File omni_vcf_index
File? gatk_bundle_tar
Int? gatk_num_cpus
Int? gatk_disk_size_gb
Int? gatk_ram_gb
Int? run_3d_dna_num_cpus
Int? run_3d_dna_disk_size_gb
Int? run_3d_dna_ram_gb
Boolean no_phasing = false
# Only for testing purposes
Boolean no_bundle = false

String docker = "encodedcc/hic-pipeline:1.11.2"
String singularity = "docker://encodedcc/hic-pipeline:1.11.2"
String docker = "encodedcc/hic-pipeline:1.11.3"
String singularity = "docker://encodedcc/hic-pipeline:1.11.3"
}

RuntimeEnvironment runtime_environment = {
Expand Down Expand Up @@ -66,15 +57,7 @@ workflow genophase {
reference_fasta_index = create_reference_fasta_index.fasta_index,
sequence_dictionary = create_gatk_references.sequence_dictionary,
interval_list = create_gatk_references.interval_list,
mills_vcf = mills_vcf,
omni_vcf = omni_vcf,
hapmap_vcf = hapmap_vcf,
dbsnp_vcf = dbsnp_vcf,
mills_vcf_index = mills_vcf_index,
omni_vcf_index = omni_vcf_index,
hapmap_vcf_index = hapmap_vcf_index,
dbsnp_vcf_index = dbsnp_vcf_index,
no_bundle = no_bundle,
bundle_tar = gatk_bundle_tar,
num_cpus = gatk_num_cpus,
ram_gb = gatk_ram_gb,
disk_size_gb = gatk_disk_size_gb,
Expand Down Expand Up @@ -154,23 +137,15 @@ task create_gatk_references {
}
}


task gatk {
input {
File bam
File reference_fasta
File reference_fasta_index
File sequence_dictionary
File interval_list
File mills_vcf
File omni_vcf
File hapmap_vcf
File dbsnp_vcf
File mills_vcf_index
File omni_vcf_index
File hapmap_vcf_index
File dbsnp_vcf_index
# Only for testing purposes
Boolean no_bundle = false
File? bundle_tar
Int num_cpus = 16
Int ram_gb = 128
Int disk_size_gb = 1000
Expand All @@ -182,25 +157,16 @@ task gatk {

command <<<
mkdir bundle
if [[ ~{if(no_bundle) then "0" else "1"} -eq 1 ]]
if [[ ~{if defined(bundle_tar) then "1" else "0"} -eq 1 ]]
then
mv \
~{mills_vcf} \
~{omni_vcf} \
~{hapmap_vcf} \
~{dbsnp_vcf} \
~{mills_vcf_index} \
~{omni_vcf_index} \
~{hapmap_vcf_index} \
~{dbsnp_vcf_index} \
bundle
tar -xvzf ~{bundle_tar} -C bundle
fi
mkdir reference
mv ~{reference_fasta_index} ~{sequence_dictionary} ~{interval_list} reference
gzip -dc ~{reference_fasta} > reference/~{basename(reference_fasta, ".gz")}
run-gatk-after-juicer2.sh \
-r reference/~{basename(reference_fasta, ".gz")} \
~{if !no_bundle then "--gatk-bundle bundle" else ""} \
~{if defined(bundle_tar) then "--gatk-bundle bundle" else ""} \
--threads ~{num_cpus} \
~{bam}
gzip -n ~{final_snp_vcf_name}
Expand Down
14 changes: 7 additions & 7 deletions hic.wdl
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -19,9 +19,9 @@ struct RuntimeEnvironment {

workflow hic {
meta {
version: "1.11.2"
caper_docker: "encodedcc/hic-pipeline:1.11.2"
caper_singularity: "docker://encodedcc/hic-pipeline:1.11.2"
version: "1.11.3"
caper_docker: "encodedcc/hic-pipeline:1.11.3"
caper_singularity: "docker://encodedcc/hic-pipeline:1.11.3"
croo_out_def: "https://raw.githubusercontent.com/ENCODE-DCC/hic-pipeline/dev/croo_out_def.json"
description: "ENCODE Hi-C pipeline, see https://github.com/ENCODE-DCC/hic-pipeline for details."
}
Expand Down Expand Up @@ -65,10 +65,10 @@ workflow hic {
Int? create_accessibility_track_disk_size_gb
String assembly_name = "undefined"

String docker = "encodedcc/hic-pipeline:1.11.2"
String singularity = "docker://encodedcc/hic-pipeline:1.11.2"
String delta_docker = "encodedcc/hic-pipeline:1.11.2_delta"
String hiccups_docker = "encodedcc/hic-pipeline:1.11.2_hiccups"
String docker = "encodedcc/hic-pipeline:1.11.3"
String singularity = "docker://encodedcc/hic-pipeline:1.11.3"
String delta_docker = "encodedcc/hic-pipeline:1.11.3_delta"
String hiccups_docker = "encodedcc/hic-pipeline:1.11.3_hiccups"
}

RuntimeEnvironment runtime_environment = {
Expand Down
2 changes: 1 addition & 1 deletion hic_pipeline/__init__.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,5 +1,5 @@
__title__ = "hic-pipeline"
__version__ = "1.11.2"
__version__ = "1.11.3"
__description__ = "ENCODE Hi-C uniform processing pipeline."
__url__ = "https://github.com/ENCODE-DCC/hic-pipeline"
__uri__ = __url__
Expand Down
10 changes: 5 additions & 5 deletions make_restriction_site_locations.wdl
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -7,9 +7,9 @@ struct RuntimeEnvironment {

workflow make_restriction_site_locations {
meta {
version: "1.11.2"
caper_docker: "encodedcc/hic-pipeline:1.11.2"
caper_singularity: "docker://encodedcc/hic-pipeline:1.11.2"
version: "1.11.3"
caper_docker: "encodedcc/hic-pipeline:1.11.3"
caper_singularity: "docker://encodedcc/hic-pipeline:1.11.3"
}

parameter_meta {
Expand All @@ -22,8 +22,8 @@ workflow make_restriction_site_locations {
File reference_fasta
String assembly_name
String restriction_enzyme
String docker = "encodedcc/hic-pipeline:1.11.2"
String singularity = "docker://encodedcc/hic-pipeline:1.11.2"
String docker = "encodedcc/hic-pipeline:1.11.3"
String singularity = "docker://encodedcc/hic-pipeline:1.11.3"
}


Expand Down
9 changes: 0 additions & 9 deletions tests/functional/json/test_genophase.json
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2,16 +2,7 @@
"genophase.bams": [
"tests/data/sample_subsampled.bam"
],
"genophase.dbsnp_vcf": "tests/data/dummy.txt",
"genophase.dbsnp_vcf_index": "tests/data/dummy.txt",
"genophase.gatk_num_cpus": 1,
"genophase.hapmap_vcf": "tests/data/dummy.txt",
"genophase.hapmap_vcf_index": "tests/data/dummy.txt",
"genophase.mills_vcf": "tests/data/dummy.txt",
"genophase.mills_vcf_index": "tests/data/dummy.txt",
"genophase.no_bundle": true,
"genophase.no_phasing": true,
"genophase.omni_vcf": "tests/data/dummy.txt",
"genophase.omni_vcf_index": "tests/data/dummy.txt",
"genophase.reference_fasta": "https://www.encodeproject.org/files/GRCh38_no_alt_analysis_set_GCA_000001405.15/@@download/GRCh38_no_alt_analysis_set_GCA_000001405.15.fasta.gz"
}