port more processes from WDL

modules/apply_recalibration/apply_recalibration.nf

@@ -0,0 +1,50 @@
+nextflow.enable.dsl = 2
+
+params.recalibrated_vcf_filename
+params.indel_filter_level
+params.snp_filter_level
+params.use_allele_specific_annotations
+
+process GATK_APPLY_RECALIBRATION {
+    container "us.gcr.io/broad-gatk/gatk:4.1.1.0"
+
+    input:
+    path(input_vcf)
+    path(input_vcf_index)
+    path(indels_recalibration)
+    path(indels_recalibration_index)
+    path(indels_tranches)
+    path(snps_recalibration)
+    path(snps_recalibration_index)
+    path(snps_tranches)
+
+
+    output:
+    path("${params.recalibrated_vcf_filename}")
+    path("${params.recalibrated_vcf_filename}.tbi")
+
+    script:
+    """
+    set -euo pipefail
+
+    gatk --java-options -Xms5g ApplyVQSR \
+                               -O tmp.indel.recalibrated.vcf \
+                               -V ${input_vcf} \
+                               --recal-file ${indels_recalibration} \
+                               ${params.use_allele_specific_annotations ? '--use-allele-specific-annotations' : ''} \
+                               --tranches-file ${indels_tranches} \
+                               --truth-sensitivity-filter-level ${indel_filter_level} \
+                               --create-output-variant-index true \
+                               -mode INDEL
+
+    gatk --java-options -Xms5g ApplyVQSR \
+                               -O ${recalibrated_vcf_filename} \
+                               -V tmp.indel.recalibrated.vcf \
+                               --recal-file ${snps_recalibration} \
+                               ${params.use_allele_specific_annotations ? '--use-allele-specific-annotations' : ''} \
+                               --tranches-file ${snps_tranches} \
+                               --truth-sensitivity-filter-level ${snp_filter_level} \
+                               --create-output-variant-index true \
+                               -mode SNP
+    """
+}

modules/collect_variant_calling_metrics/collect_variant_calling_metrics.nf

@@ -0,0 +1,30 @@
+nextflow.enable.dsl = 2
+
+process GATK_COLLECT_VARIANT_CALLING_METRICS {
+
+    input:
+    path(input_vcf)
+    path(input_vcf_index)
+    path(dbsnp_vcf)
+    path(dbsnp_vcf_index)
+    path(interval_list)
+    path(ref_dict)
+
+    output:
+    path("${params.metrics_file_prefix}.variant_calling_detail_metrics")
+    path("${params.metrics_file_prefix}.variant_calling_summary_metrics")
+
+    script:
+    """
+    set -euo pipefail
+
+    gatk --java-options -Xms6g CollectVariantCallingMetrics \
+                               --INPUT ${input_vcf} \
+                               --DBSNP ${dbsnp_vcf} \
+                               --SEQUENCE_DICTIONARY ${ref_dict} \
+                               --OUTPUT ${params.metrics_filename_prefix} \
+                               --THREAD_COUNT 8 \
+                               --TARGET_INTERVALS ${interval_list}
+
+    """
+}

modules/gather_tranches/gather_tranches.nf

@@ -0,0 +1,52 @@
+nextflow.enable.dsl = 2
+
+params.output_filename
+
+process GATK_GATHER_TRANCHES {
+    container "us.gcr.io/broad-gotc-prod/gatk4-joint-genotyping:1.3.0-1527875152"
+
+    input:
+    path(tranches)
+
+
+    output:
+    path("${params.output_filename}")
+
+
+    shell:
+//    NOTE: This assumes that gsutil is installed
+    tranches_lines = write_lines(tranches)
+
+    '''
+    set -euo pipefail
+
+    tranches_fofn=!{tranches_lines}
+
+    # Jose says:
+    # Cromwell will fall over if we have it try to localize tens of thousands of files,
+    # so we manually localize files using gsutil.
+    # Using gsutil also lets us parallelize the localization, which (as far as we can tell)
+    # PAPI doesn't do.
+
+    # This is here to deal with the JES bug where commands may be run twice
+    rm -rf tranches
+    mkdir tranches
+    RETRY_LIMIT=5
+
+    count=0
+    until cat $tranches_fofn | /root/google-cloud-sdk/bin/gsutil -m cp -L cp.log -c -I tranches/; do
+    sleep 1
+    ((count++)) && ((count >= $RETRY_LIMIT)) && break
+    done
+    if [ "$count" -ge "$RETRY_LIMIT" ]; then
+    echo 'Could not copy all the tranches from the cloud' && exit 1
+    fi
+
+    cat $tranches_fofn | rev | cut -d '/' -f 1 | rev | awk '{print "tranches/" $1}' > inputs.list
+
+    /usr/gitc/gatk --java-options -Xms6g GatherTranches \
+                                        --input inputs.list \
+                                        --output !{output_filename}
+    '''
+
+}

modules/gather_variant_calling_metrics/gather_variant_calling_metrics.nf

@@ -0,0 +1,5 @@
+nextflow.enable.dsl = 2
+
+process GATK_GATHER_VARIANT_CALLING_METRICS {
+
+}

modules/gather_vcfs/gather_vcfs.nf

@@ -0,0 +1,29 @@
+nextflow.enable.dsl = 2
+
+process GATK_GATHER_VCFS {
+    container "us.gcr.io/broad-gatk/gatk:4.1.1.0"
+
+    output:
+    path("${params.output_vcf_name}")
+    path("${params.output_vcf_name}.tbi")
+
+    script:
+
+    input_vcfs_str = input_vcfs.collect {" --input "}
+
+    """
+    set -euo pipefail
+
+    # --ignore-safety-checks makes a big performance difference so we include it in our invocation.
+    # This argument disables expensive checks that the file headers contain the same set of
+    # genotyped samples and that files are in order by position of first record.
+    gatk --java-options -Xms6g GatherVcfsCloud \
+                               --ignore-safety-checks \
+                               --gather-type BLOCK \
+                               --input ${input_vcfs_str} \
+                               --output ${params.output_vcf_name}
+
+    tabix ${params.output_vcf_name}
+   """
+
+}

modules/indels_variant_recalibrator/indels_variant_recalibrator.nf

@@ -7,6 +7,7 @@ params.use_allele_specific_annotations
 
 
 process GATK_INDELS_VARIANT_RECALIBRATOR {
+    container "us.gcr.io/broad-gatk/gatk:4.1.1.0"
 
     input:
     path(sites_only_variant_filtered_vcf)

modules/select_fingerprint_site_variants/select_fingerprint_site_variants.nf

@@ -0,0 +1,32 @@
+nextflow.enable.dsl = 2
+
+params.base_output_name
+
+process GATK_SELECT_FINGERPRINT_SITE_VARIANTS {
+
+    input:
+    path(input_vcf)
+    path(haplotype_database)
+
+    output:
+    path("${params.base_output_name}.vcf.gz")
+    path("${params.base_output_name}.vcf.gz.tbi")
+
+    shell:
+
+    '''
+    set -euo pipefail
+
+    function hdb_to_interval_list() {
+        input=$1
+        awk 'BEGIN{IFS="\t";OFS="\t";} $0~"^@"{print;next;} $0~"#CHROM"{next;} {print $1,$2,$2,"+","interval-"NR}' $1
+    }
+
+    hdb_to_interval_list !{haplotype_database} > hdb.interval_list
+
+    gatk --java-options -Xms6g SelectVariants \
+                        --variant !{input_vcf} \
+                        --intervals hdb.interval_list \
+                        --output !{params.base_output_name}.vcf.gz
+    '''
+}

modules/snps_variant_recalibrator/snps_variant_recalibrator.nf

@@ -0,0 +1,55 @@
+nextflow.enable.dsl = 2
+
+params.model_report_filename
+pararms.recalibration_filename
+params.tranches_filename
+params.use_allele_specific_annotations
+
+process GATK_SNPS_VARIANT_RECALIBRATOR {
+    container "us.gcr.io/broad-gatk/gatk:4.1.1.0"
+
+    input:
+    path(sites_only_variant_filtered_vcf)
+    path(sites_only_variant_filtered_vcf_index)
+    path(hapmap_resource_vcf)
+    path(omni_resource_vcf)
+    path(one_thousand_genomes_resource_vcf)
+    path(dbsnp_resource_vcf)
+    path(hapmap_resource_vcf_index)
+    path(omni_resource_vcf_index)
+    path(one_thousand_genomes_resource_vcf_index)
+    path(dbsnp_resource_vcf_index)
+
+    output:
+    path("${params.recalibration_filename}")
+    path("${params.recalibration_filename}.idx")
+    path("${params.tranches_filename}")
+
+
+    shell:
+    tranche_str = recalibration_tranche_values.collect().join(" -tranche ")
+    an_str = recalibration_annotation_values.collect().join(" -an ")
+
+    '''
+    set -euo pipefail
+
+    MODEL_REPORT=!{model_report}
+
+    gatk --java-options -Xms!{java_mem}g  VariantRecalibrator \
+                                          -V !{sites_only_variant_filtered_vcf} \
+                                          -O !{recalibration_filename} \
+                                          --tranches-file !{tranches_filename} \
+                                          --trust-all-polymorphic \
+                                          -tranche !{tranche_str} \
+                                          -an !{an_str} \
+                                          !{params.use_allele_specific_annotations ? '--use-allele-specific-annotations' : ''} \
+                                          -mode SNP \
+                                          !{model_report_arg} \
+                                          --max-gaussians !{max_gaussians} \
+                                          -resource:hapmap,known=false,training=true,truth=true,prior=15 !{hapmap_resource_vcf} \
+                                          -resource:omni,known=false,training=true,truth=true,prior=12 !{omni_resource_vcf} \
+                                          -resource:1000G,known=false,training=true,truth=false,prior=10 !{one_thousand_genomes_resource_vcf} \
+                                          -resource:dbsnp,known=true,training=false,truth=false,prior=7 !{dbsnp_resource_vcf}
+    
+    '''
+}

modules/snps_variant_recalibrator_create_model/snps_variant_recalibrator_create_model.nf

@@ -6,6 +6,7 @@ params.tranches_filename
 params.use_allele_specific_annotations
 
 process GATK_SNPS_VARIANT_RECALIBRATOR_CREATE_MODEL {
+    container "us.gcr.io/broad-gatk/gatk:4.1.1.0"
 
     input:
     path(sites_only_variant_filtered_vcf)