Convert WDL command blocks from brace to heredoc syntax

Update all WDL task files to use modern `command <<<` heredoc syntax
instead of the older `command {` brace style. This ensures consistent
behavior with shell special characters (quotes, braces) across all
WDL interpreters.

Also updates legacy `${var}` variable interpolation to `~{var}` syntax
in affected tasks.

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

Author: dpark01

pipes/WDL/tasks/tasks_assembly.wdl

@@ -60,7 +60,7 @@ task assemble {
 
     Int disk_size = 375
 
-    command {
+    command <<<
         set -ex -o pipefail
 
         # find 90% memory
@@ -87,7 +87,7 @@ task assemble {
         cat /proc/loadavg | cut -f 3 -d ' ' > LOAD_15M
         set +o pipefail
         { if [ -f /sys/fs/cgroup/memory.peak ]; then cat /sys/fs/cgroup/memory.peak; elif [ -f /sys/fs/cgroup/memory/memory.peak ]; then cat /sys/fs/cgroup/memory/memory.peak; elif [ -f /sys/fs/cgroup/memory/memory.max_usage_in_bytes ]; then cat /sys/fs/cgroup/memory/memory.max_usage_in_bytes; else echo "0"; fi; } > MEM_BYTES
-    }
+    >>>
 
     output {
         File   contigs_fasta        = "~{sample_name}.assembly1-spades.fasta"
@@ -574,7 +574,7 @@ task ivar_trim {
       }
     }
 
-    command {
+    command <<<
         ivar version | head -1 | tee VERSION
         if [ -f "~{trim_coords_bed}" ]; then
           ivar trim -e \
@@ -593,7 +593,7 @@ task ivar_trim {
         PCT=$(grep "Trimmed primers from" IVAR_OUT | perl -lape 's/Trimmed primers from (\S+)%.*/$1/')
         if [[ $PCT = -* ]]; then echo 0; else echo $PCT; fi > IVAR_TRIM_PCT
         grep "Trimmed primers from" IVAR_OUT | perl -lape 's/Trimmed primers from \S+% \((\d+)\).*/$1/' > IVAR_TRIM_COUNT
-    }
+    >>>
 
     output {
         File   aligned_trimmed_bam         = "~{bam_basename}.trimmed.bam"
@@ -992,7 +992,7 @@ task run_discordance {
 
     Int disk_size = 100
 
-    command {
+    command <<<
         set -ex -o pipefail
 
         read_utils.py --version | tee VERSION
@@ -1058,7 +1058,7 @@ task run_discordance {
           echo 0 > num_discordant_snps
           echo 0 > num_discordant_indels
         fi
-    }
+    >>>
 
     output {
         File   discordant_sites_vcf = "~{out_basename}.discordant.vcf"

pipes/WDL/tasks/tasks_demux.wdl

@@ -11,7 +11,7 @@ task merge_tarballs {
 
   Int disk_size = 2625
 
-  command {
+  command <<<
     set -ex -o pipefail
 
     if [ -z "$TMPDIR" ]; then
@@ -23,7 +23,7 @@ task merge_tarballs {
     file_utils.py merge_tarballs \
       ~{out_filename} ~{sep=' ' tar_chunks} \
       --loglevel=DEBUG
-  }
+  >>>
 
   output {
     File   combined_tar     = "~{out_filename}"

pipes/WDL/tasks/tasks_intrahost.wdl

@@ -202,24 +202,24 @@ task isnvs_per_sample {
   }
 
 
-  command {
+  command <<<
     intrahost.py --version | tee VERSION
     intrahost.py vphaser_one_sample \
-        ${mapped_bam} \
-        ${assembly_fasta} \
-        vphaser2.${sample_name}.txt.gz \
-        ${'--vphaserNumThreads=' + threads} \
-        ${true="--removeDoublyMappedReads" false="" removeDoublyMappedReads} \
-        ${'--minReadsEach=' + minReadsPerStrand} \
-        ${'--maxBias=' + maxBias}
-  }
+        ~{mapped_bam} \
+        ~{assembly_fasta} \
+        vphaser2.~{sample_name}.txt.gz \
+        ~{'--vphaserNumThreads=' + threads} \
+        ~{true="--removeDoublyMappedReads" false="" removeDoublyMappedReads} \
+        ~{'--minReadsEach=' + minReadsPerStrand} \
+        ~{'--maxBias=' + maxBias}
+  >>>
 
   output {
-    File   isnvsFile        = "vphaser2.${sample_name}.txt.gz"
+    File   isnvsFile        = "vphaser2.~{sample_name}.txt.gz"
     String viralngs_version = read_string("VERSION")
   }
   runtime {
-    docker: "${docker}"
+    docker: docker
     memory: select_first([machine_mem_gb, 7]) + " GB"
     dx_instance_type: "mem1_ssd1_v2_x8"
     maxRetries: 2
@@ -250,43 +250,43 @@ task isnvs_vcf {
     emailAddress:              { description: "email address passed to NCBI if we need to download reference sequences" }
   }
 
-  command {
+  command <<<
     set -ex -o pipefail
 
     intrahost.py --version | tee VERSION
 
-    SAMPLES="${sep=' ' sampleNames}"
+    SAMPLES="~{sep=' ' sampleNames}"
     if [ -n "$SAMPLES" ]; then SAMPLES="--samples $SAMPLES"; fi
 
-    providedSnpRefAccessions="${sep=' ' snpEffRef}"
-    if [ -n "$providedSnpRefAccessions" ]; then 
+    providedSnpRefAccessions="~{sep=' ' snpEffRef}"
+    if [ -n "$providedSnpRefAccessions" ]; then
       snpRefAccessions="$providedSnpRefAccessions";
     else
-      snpRefAccessions="$(python -c "from Bio import SeqIO; print(' '.join(list(s.id for s in SeqIO.parse('${reference_fasta}', 'fasta'))))")"
+      snpRefAccessions="$(python -c "from Bio import SeqIO; print(' '.join(list(s.id for s in SeqIO.parse('~{reference_fasta}', 'fasta'))))")"
     fi
 
     echo "snpRefAccessions: $snpRefAccessions"
 
     intrahost.py merge_to_vcf \
-        ${reference_fasta} \
+        ~{reference_fasta} \
         isnvs.vcf.gz \
         $SAMPLES \
-        --isnvs ${sep=' ' vphaser2Calls} \
-        --alignments ${sep=' ' perSegmentMultiAlignments} \
+        --isnvs ~{sep=' ' vphaser2Calls} \
+        --alignments ~{sep=' ' perSegmentMultiAlignments} \
         --strip_chr_version \
-        ${true="--naive_filter" false="" naiveFilter} \
+        ~{true="--naive_filter" false="" naiveFilter} \
         --parse_accession
-        
+
     interhost.py snpEff \
         isnvs.vcf.gz \
         $snpRefAccessions \
         isnvs.annot.vcf.gz \
-        ${'--emailAddress=' + emailAddress}
+        ~{'--emailAddress=' + emailAddress}
 
     intrahost.py iSNV_table \
         isnvs.annot.vcf.gz \
         isnvs.annot.txt.gz
-  }
+  >>>
 
   output {
     File   isnvs_vcf           = "isnvs.vcf.gz"
@@ -297,7 +297,7 @@ task isnvs_vcf {
     String viralngs_version    = read_string("VERSION")
   }
   runtime {
-    docker: "${docker}"
+    docker: docker
     memory: select_first([machine_mem_gb, 4]) + " GB"
     dx_instance_type: "mem1_ssd1_v2_x4"
     maxRetries: 2
@@ -328,27 +328,27 @@ task annotate_vcf_snpeff {
     emailAddress:       { description: "email address passed to NCBI if we need to download reference sequences" }
   }
 
-  command {
+  command <<<
     set -ex -o pipefail
 
     intrahost.py --version | tee VERSION
 
-    providedSnpRefAccessions="${sep=' ' snpEffRef}"
-    if [ -n "$providedSnpRefAccessions" ]; then 
+    providedSnpRefAccessions="~{sep=' ' snpEffRef}"
+    if [ -n "$providedSnpRefAccessions" ]; then
       snpRefAccessions="$providedSnpRefAccessions";
     else
-      snpRefAccessions="$(python -c "from Bio import SeqIO; print(' '.join(list(s.id for s in SeqIO.parse('${ref_fasta}', 'fasta'))))")"
+      snpRefAccessions="$(python -c "from Bio import SeqIO; print(' '.join(list(s.id for s in SeqIO.parse('~{ref_fasta}', 'fasta'))))")"
     fi
     echo "snpRefAccessions: $snpRefAccessions"
 
     vcf_to_use=""
-    if (file "${in_vcf}" | grep -q "gzip" ) ; then
-      echo "${in_vcf} is already compressed"
-      vcf_to_use="${in_vcf}"
+    if (file "~{in_vcf}" | grep -q "gzip" ) ; then
+      echo "~{in_vcf} is already compressed"
+      vcf_to_use="~{in_vcf}"
     else
-      echo "${in_vcf} is not compressed; gzipping..."
-      bgzip "${in_vcf}"
-      vcf_to_use="${in_vcf}.gz"
+      echo "~{in_vcf} is not compressed; gzipping..."
+      bgzip "~{in_vcf}"
+      vcf_to_use="~{in_vcf}.gz"
     fi
 
     # renames the seq id using the first sequence in the alignment
@@ -360,7 +360,7 @@ task annotate_vcf_snpeff {
     bgzip -d "temp.vcf.gz"
     # rename chr field (first col) in vcf
     cat "temp.vcf" | sed "s/^1/$ref_name_no_version/" > "temp2.vcf"
-    
+
     # output the vcf, removing the reference sequence if present as a sample name
     bgzip "temp2.vcf"
     tabix -p vcf "temp2.vcf.gz"
@@ -377,13 +377,13 @@ task annotate_vcf_snpeff {
     echo "Creating vcf index"
     bcftools index "$vcf_to_use"
     tabix -p vcf "$vcf_to_use"
-    
+
     interhost.py snpEff \
         "$vcf_to_use" \
         $snpRefAccessions \
-        "${output_basename}.annot.vcf.gz" \
-        ${'--emailAddress=' + emailAddress}
-  }
+        "~{output_basename}.annot.vcf.gz" \
+        ~{'--emailAddress=' + emailAddress}
+  >>>
 
   output {
     File   annot_vcf_gz     = "~{output_basename}.annot.vcf.gz"

pipes/WDL/tasks/tasks_ncbi.wdl

@@ -9,17 +9,17 @@ task download_fasta {
     String         docker = "quay.io/broadinstitute/viral-phylo:2.5.16.0"
   }
 
-  command {
+  command <<<
     ncbi.py --version | tee VERSION
     ncbi.py fetch_fastas \
-        ${emailAddress} \
+        ~{emailAddress} \
         . \
-        ${sep=' ' accessions} \
-        --combinedFilePrefix ${out_prefix} \
-  }
+        ~{sep=' ' accessions} \
+        --combinedFilePrefix ~{out_prefix} \
+  >>>
 
   output {
-    File   sequences_fasta  = "${out_prefix}.fasta"
+    File   sequences_fasta  = "~{out_prefix}.fasta"
     String viralngs_version = read_string("VERSION")
   }
 
@@ -467,11 +467,11 @@ task rename_fasta_header {
 
     String docker = "quay.io/broadinstitute/viral-core:2.5.20"
   }
-  command {
+  command <<<
     set -e
     file_utils.py rename_fasta_sequences \
       "~{genome_fasta}" "~{out_basename}.fasta" "~{new_name}"
-  }
+  >>>
   output {
     File renamed_fasta = "~{out_basename}.fasta"
   }
@@ -600,10 +600,10 @@ task lookup_table_by_filename {
 
     String docker = "ubuntu"
   }
-  command {
+  command <<<
     set -e -o pipefail
     grep ^"~{id}" ~{mapping_tsv} | cut -f ~{return_col} > OUTVAL
-  }
+  >>>
   output {
     String value = read_string("OUTVAL")
   }

pipes/WDL/tasks/tasks_nextstrain.wdl

@@ -649,11 +649,11 @@ task nextstrain_ncov_defaults {
         String docker                      = "docker.io/nextstrain/base:build-20240318T173028Z"
         Int    disk_size = 50
     }
-    command {
+    command <<<
         set -e
         wget -q "https://github.com/nextstrain/ncov/archive/~{nextstrain_ncov_repo_commit}.tar.gz"
         tar -xf "~{nextstrain_ncov_repo_commit}.tar.gz" --strip-components=1
-    }
+    >>>
     runtime {
         docker: docker
         memory: "1 GB"
@@ -700,7 +700,7 @@ task nextstrain_deduplicate_sequences {
 
     String out_basename = basename(basename(basename(basename(sequences_fasta, '.xz'), '.gz'), '.tar'), '.fasta')
     String out_filename = "~{out_basename}_sequences_deduplicated.fasta"
-    command {
+    command <<<
         set -e
         ncov_path_prefix="/nextstrain/ncov"
         wget -q "https://github.com/nextstrain/ncov/archive/~{nextstrain_ncov_repo_commit}.tar.gz"
@@ -709,9 +709,9 @@ task nextstrain_deduplicate_sequences {
 
         python3 "$ncov_path_prefix/scripts/sanitize_sequences.py" \
         --sequences "~{sequences_fasta}" \
-        ${true="--error-on-duplicate-strains" false="" error_on_seq_diff} \
+        ~{true="--error-on-duplicate-strains" false="" error_on_seq_diff} \
         --output "~{out_filename}"
-    }
+    >>>
     runtime {
         docker: docker
         memory: "7 GB"
@@ -757,7 +757,7 @@ task nextstrain_ncov_sanitize_gisaid_data {
     }
 
     String out_basename = basename(basename(basename(basename(sequences_gisaid_fasta, '.xz'), '.gz'), '.tar'), '.fasta')
-    command {
+    command <<<
         set -e
         ncov_path_prefix="/nextstrain/ncov"
         wget -q "https://github.com/nextstrain/ncov/archive/~{nextstrain_ncov_repo_commit}.tar.gz"
@@ -775,7 +775,7 @@ task nextstrain_ncov_sanitize_gisaid_data {
         --rename-fields 'Virus name=strain' 'Accession ID=gisaid_epi_isl' 'Collection date=date' 'Clade=GISAID_clade' 'Pango lineage=pango_lineage' 'Host=host' 'Type=virus' 'Patient age=age' \
         ~{"--strip-prefixes=" + prefix_to_strip} \
         --output "~{out_basename}_metadata_sanitized_for_nextstrain.tsv.gz"
-    }
+    >>>
     runtime {
         docker: docker
         memory: "7 GB"
@@ -1246,10 +1246,10 @@ task snp_sites {
         Int     disk_size = 750
     }
     String out_basename = basename(msa_fasta, ".fasta")
-    command {
+    command <<<
         snp-sites -V > VERSION
         snp-sites -v ~{true="" false="-c" allow_wildcard_bases} -o "~{out_basename}.vcf" "~{msa_fasta}"
-    }
+    >>>
     runtime {
         docker: docker
         memory: "31 GB"

pipes/WDL/tasks/tasks_read_utils.wdl

@@ -426,32 +426,32 @@ task downsample_bams {
 
   Int disk_size = 750
 
-  command {
+  command <<<
     set -ex -o pipefail
 
     # find 90% memory
     mem_in_mb=$(/opt/viral-ngs/source/docker/calc_mem.py mb 90)
 
-    if [[ "${deduplicateBefore}" == "true" ]]; then
+    if [[ "~{deduplicateBefore}" == "true" ]]; then
       DEDUP_OPTION="--deduplicateBefore"
-    elif [[ "${deduplicateAfter}" == "true" ]]; then
+    elif [[ "~{deduplicateAfter}" == "true" ]]; then
       DEDUP_OPTION="--deduplicateAfter"
     fi
 
-    if [[ "${deduplicateBefore}" == "true" && "${deduplicateAfter}" == "true" ]]; then
+    if [[ "~{deduplicateBefore}" == "true" && "~{deduplicateAfter}" == "true" ]]; then
       echo "deduplicateBefore and deduplicateAfter are mutually exclusive. Only one can be used."
       exit 1
     fi
-    
+
     read_utils.py --version | tee VERSION
 
     read_utils.py downsample_bams \
-        ${sep=' ' reads_bam} \
+        ~{sep=' ' reads_bam} \
         --outPath ./output \
-        ${'--readCount=' + readCount} \
+        ~{'--readCount=' + readCount} \
         $DEDUP_OPTION \
         --JVMmemory "$mem_in_mb"m
-  }
+  >>>
 
   output {
     Array[File] downsampled_bam  = glob("output/*.downsampled-*.bam")