de-icelake a bunch of tasks since it seems unnecessary

broadinstitute · Mar 1, 2024 · d7152d7 · d7152d7
1 parent 8696c3f
commit d7152d7
Show file tree

Hide file tree

Showing 6 changed files with 15 additions and 38 deletions.
diff --git a/pipes/WDL/tasks/tasks_demux.wdl b/pipes/WDL/tasks/tasks_demux.wdl
@@ -39,7 +39,6 @@ task merge_tarballs {
     dx_instance_type: "mem1_ssd2_v2_x16"
     maxRetries: 2
     preemptible: 0
-    cpuPlatform: "Intel Ice Lake"
   }
 }
 
@@ -428,7 +427,6 @@ task illumina_demux {
     dx_timeout: "20H"
     maxRetries: 2
     preemptible: 0  # this is the very first operation before scatter, so let's get it done quickly & reliably
-    cpuPlatform: "Intel Ice Lake"
   }
 }
 

diff --git a/pipes/WDL/tasks/tasks_metagenomics.wdl b/pipes/WDL/tasks/tasks_metagenomics.wdl
@@ -127,7 +127,6 @@ task krakenuniq {
     dx_instance_type: "mem3_ssd1_v2_x48"
     preemptible: 0
     maxRetries: 2
-    cpuPlatform: "Intel Ice Lake"
   }
 }
 
@@ -198,7 +197,6 @@ task build_krakenuniq_db {
     dx_instance_type: "mem3_ssd1_v2_x32"
     preemptible: 0
     maxRetries: 2
-    cpuPlatform: "Intel Ice Lake"
   }
 }
 
@@ -319,13 +317,12 @@ task kraken2 {
   runtime {
     docker: docker
     memory: machine_mem_gb + " GB"
-    cpu: 8
+    cpu: 16
     disks:  "local-disk " + disk_size + " LOCAL"
     disk: disk_size + " GB" # TESs
     dx_instance_type: "mem3_ssd1_v2_x8"
     preemptible: 2
     maxRetries: 2
-    cpuPlatform: "Intel Ice Lake"
   }
 }
 
@@ -567,7 +564,6 @@ task build_kraken2_db {
     dx_instance_type: "mem3_ssd1_v2_x16"
     preemptible: 0
     maxRetries: 2
-    cpuPlatform: "Intel Ice Lake"
   }
 }
 
@@ -656,7 +652,6 @@ task blastx {
     dx_instance_type: "mem1_ssd1_v2_x36"
     preemptible: 1
     maxRetries: 2
-    cpuPlatform: "Intel Ice Lake"
   }
 }
 
@@ -856,7 +851,6 @@ task filter_bam_to_taxa {
     cpu: 8
     dx_instance_type: "mem3_ssd1_v2_x4"
     maxRetries: 2
-    cpuPlatform: "Intel Ice Lake"
   }
 }
 
@@ -933,6 +927,5 @@ task kaiju {
     disk: disk_size + " GB" # TES
     dx_instance_type: "mem3_ssd1_v2_x16"
     maxRetries: 2
-    cpuPlatform: "Intel Ice Lake"
   }
 }
diff --git a/pipes/WDL/tasks/tasks_nextstrain.wdl b/pipes/WDL/tasks/tasks_nextstrain.wdl
@@ -576,7 +576,6 @@ task nextstrain_build_subsample {
         disk: disk_size + " GB" # TES
         dx_instance_type: "mem3_ssd1_v2_x8"
         maxRetries: 2
-        cpuPlatform: "Intel Ice Lake"
     }
     output {
         File            subsampled_msa  = "ncov/results/~{build_name}/~{build_name}_subsampled_sequences.fasta"
@@ -669,7 +668,6 @@ task nextstrain_deduplicate_sequences {
         disk: disk_size + " GB" # TES
         dx_instance_type: "mem2_ssd1_v2_x2"
         maxRetries: 2
-        cpuPlatform: "Intel Ice Lake"
     }
     output {
         File sequences_deduplicated_fasta = "~{out_filename}"
@@ -828,7 +826,6 @@ task filter_subsample_sequences {
         dx_instance_type: "mem1_ssd1_v2_x4"
         preemptible: 1
         maxRetries: 2
-        cpuPlatform: "Intel Ice Lake"
     }
     output {
         File   filtered_fasta    = out_fname
@@ -927,7 +924,6 @@ task filter_sequences_to_list {
         dx_instance_type: "mem1_ssd1_v2_x4"
         preemptible: 1
         maxRetries: 2
-        cpuPlatform: "Intel Ice Lake"
     }
     output {
         File   filtered_fasta    = out_fname
@@ -1176,7 +1172,6 @@ task augur_mafft_align {
         preemptible: 0
         dx_instance_type: "mem3_ssd2_v2_x32"
         maxRetries: 2
-        cpuPlatform: "Intel Ice Lake"
     }
     output {
         File   aligned_sequences = "~{basename}_aligned.fasta"

diff --git a/pipes/WDL/tasks/tasks_read_utils.wdl b/pipes/WDL/tasks/tasks_read_utils.wdl
@@ -231,7 +231,6 @@ task merge_and_reheader_bams {
         dx_instance_type: "mem1_ssd2_v2_x4"
         preemptible: 0
         maxRetries: 2
-        cpuPlatform: "Intel Ice Lake"
     }
 }
 
@@ -244,7 +243,7 @@ task rmdup_ubam {
     File    reads_unmapped_bam
     String  method = "mvicuna"
 
-    Int?    machine_mem_gb
+    Int     machine_mem_gb = 7
     String  docker = "quay.io/broadinstitute/viral-core:2.2.4"
   }
 
@@ -257,38 +256,37 @@ task rmdup_ubam {
 
   String reads_basename = basename(reads_unmapped_bam, ".bam")
   
-  command {
+  command <<<
     set -ex -o pipefail
     mem_in_mb=$(/opt/viral-ngs/source/docker/calc_mem.py mb 90)
     read_utils.py --version | tee VERSION
 
-    read_utils.py rmdup_"${method}"_bam \
-      "${reads_unmapped_bam}" \
-      "${reads_basename}".dedup.bam \
+    read_utils.py rmdup_"~{method}"_bam \
+      "~{reads_unmapped_bam}" \
+      "~{reads_basename}".dedup.bam \
       --JVMmemory "$mem_in_mb"m \
       --loglevel=DEBUG
 
-    samtools view -c ${reads_basename}.dedup.bam | tee dedup_read_count_post
-    reports.py fastqc ${reads_basename}.dedup.bam ${reads_basename}.dedup_fastqc.html --out_zip ${reads_basename}.dedup_fastqc.zip
-  }
+    samtools view -c "~{reads_basename}.dedup.bam" | tee dedup_read_count_post
+    reports.py fastqc "~{reads_basename}.dedup.bam" "~{reads_basename}.dedup_fastqc.html" --out_zip "~{reads_basename}.dedup_fastqc.zip"
+  >>>
 
   output {
-    File   dedup_bam             = "${reads_basename}.dedup.bam"
-    File   dedup_fastqc          = "${reads_basename}.dedup_fastqc.html"
-    File   dedup_fastqc_zip      = "${reads_basename}.dedup_fastqc.zip"
+    File   dedup_bam             = "~{reads_basename}.dedup.bam"
+    File   dedup_fastqc          = "~{reads_basename}.dedup_fastqc.html"
+    File   dedup_fastqc_zip      = "~{reads_basename}.dedup_fastqc.zip"
     Int    dedup_read_count_post = read_int("dedup_read_count_post")
     String viralngs_version      = read_string("VERSION")
   }
 
   runtime {
     docker: docker
-    memory: select_first([machine_mem_gb, 7]) + " GB"
+    memory: machine_mem_gb + " GB"
     cpu:    2
     disks:  "local-disk " + disk_size + " LOCAL"
     disk: disk_size + " GB" # TES
     dx_instance_type: "mem2_ssd1_v2_x2"
     maxRetries: 2
-    cpuPlatform: "Intel Ice Lake"
   }
 }
 
@@ -301,8 +299,8 @@ task downsample_bams {
   input {
     Array[File]+ reads_bam
     Int?         readCount
-    Boolean?     deduplicateBefore = false
-    Boolean?     deduplicateAfter = false
+    Boolean      deduplicateBefore = false
+    Boolean      deduplicateAfter = false
 
     Int?         machine_mem_gb
     String       docker = "quay.io/broadinstitute/viral-core:2.2.4"
@@ -350,7 +348,6 @@ task downsample_bams {
     disk: disk_size + " GB" # TES
     dx_instance_type: "mem1_ssd1_v2_x4"
     maxRetries: 2
-    cpuPlatform: "Intel Ice Lake"
   }
 }
 
@@ -410,7 +407,6 @@ task FastqToUBAM {
     disk: disk_size + " GB" # TES
     dx_instance_type: "mem1_ssd1_v2_x2"
     maxRetries: 2
-    cpuPlatform: "Intel Ice Lake"
   }
   output {
     File unmapped_bam = "~{sample_name}.bam"

diff --git a/pipes/WDL/tasks/tasks_taxon_filter.wdl b/pipes/WDL/tasks/tasks_taxon_filter.wdl
@@ -99,7 +99,6 @@ task deplete_taxa {
     dx_instance_type: "mem1_ssd1_v2_x8"
     preemptible: 1
     maxRetries: 2
-    cpuPlatform: "Intel Ice Lake"
   }
 }
 
@@ -165,7 +164,6 @@ task filter_to_taxon {
     disk: disk_size + " GB" # TES
     dx_instance_type: "mem1_ssd1_v2_x8"
     maxRetries: 2
-    cpuPlatform: "Intel Ice Lake"
   }
 }
 
@@ -255,6 +253,5 @@ task merge_one_per_sample {
     disk: disk_size + " GB" # TES
     dx_instance_type: "mem1_ssd2_v2_x4"
     maxRetries: 2
-    cpuPlatform: "Intel Ice Lake"
   }
 }
diff --git a/pipes/WDL/tasks/tasks_utils.wdl b/pipes/WDL/tasks/tasks_utils.wdl
@@ -21,7 +21,6 @@ task concatenate {
         disk: disk_size + " GB" # TES
         dx_instance_type: "mem1_ssd1_v2_x2"
         maxRetries: 2
-        cpuPlatform: "Intel Ice Lake"
     }
     output {
         File combined = "${output_name}"
@@ -112,7 +111,6 @@ task zcat {
         disk: disk_size + " GB" # TES
         dx_instance_type: "mem1_ssd1_v2_x2"
         maxRetries: 2
-        cpuPlatform: "Intel Ice Lake"
     }
     output {
         File    combined     = "${output_name}"