Merge pull request #10 from theiagen/jro-snippy_tree-dev

Add snippy core and snippy tree workflows

.dockstore.yml

@@ -15,6 +15,11 @@ workflows:
    primaryDescriptorPath: /workflows/wf_snippy_variants.wdl
    testParameterFiles:
     - empty.json
+ - name: Snippy_Tree
+   subclass: WDL
+   primaryDescriptorPath: /workflows/wf_snippy_tree.wdl
+   testParameterFiles:
+    - empty.json
  - name: Nullarbor
    subclass: WDL
    primaryDescriptorPath: /workflows/wf_nullarbor.wdl

tasks/gene_typing/task_snippy_variants.wdl

@@ -8,8 +8,8 @@ task snippy_variants {
     String? query_gene
     String samplename
     String docker = "staphb/snippy:4.6.0"
-    Int cpus = 4
-    Int memory = 16
+    Int cpus = 8
+    Int memory = 32
     # Paramters 
     # --map_qual: Minimum read mapping quality to consider (default '60')
     # --base_quality: Minimum base quality to consider (default '13')
@@ -36,7 +36,7 @@ task snippy_variants {
     if [ -z "~{query_gene}" ]; then 
      no_hit="NA: No query gene was provided"
    else 
-    no_hit="No variants identified in quieried genes (~{query_gene})" 
+    no_hit="No variants identified in queried genes (~{query_gene})" 
     fi
     # call snippy
       snippy \
@@ -70,13 +70,15 @@ task snippy_variants {
    else
      echo "${no_hit}" > SNIPPY_VARIANT_HITS
    fi
+    # Compress output dir
+    tar -cvzf "./~{samplename}_snippy_variants_outdir.tar" "./~{samplename}"
   >>>
   output {
     String snippy_variants_version = read_string("VERSION")
     String snippy_variants_query = "~{query_gene}"
     String snippy_variants_hits = read_string("SNIPPY_VARIANT_HITS")
+    File snippy_variants_outdir_tarball = "./~{samplename}_snippy_variants_outdir.tar"
     File snippy_variants_gene_query_results = "./gene_query.csv"
-    Array[File] snippy_outputs =  glob("~{samplename}/~{samplename}*")
     File snippy_variants_results = "~{samplename}/~{samplename}.csv"
     File snippy_variants_bam = "~{samplename}/~{samplename}.bam"
     File snippy_variants_bai ="~{samplename}/~{samplename}.bam.bai"

tasks/phylogenetic_inference/task_gubbins.wdl

@@ -0,0 +1,54 @@
+version 1.0
+
+task gubbins {
+  input {
+    File alignment
+    String cluster_name
+    String docker = "sangerpathogens/gubbins"
+    Int? filter_percent = 25 #default is 25%
+    Int? iterations = 5
+    String? tree_builder = "raxml"
+    String? tree_args
+    String? nuc_subst_model = "GTRGAMMA"
+    Int? bootstrap = 0
+    String? outgroup
+    File? dates_file
+  }
+  command <<<
+    # date and version control
+    date | tee DATE
+    run_gubbins.py --version | tee VERSION
+
+    run_gubbins.py \
+    ~{alignment} \
+    --prefix ~{cluster_name} \
+    --filter-percentage ~{filter_percent} \
+    --iterations ~{iterations} \
+    --tree-builder ~{tree_builder} \
+    ~{'--tree-args ' + tree_args} \
+    ~{'--model ' + nuc_subst_model} \
+    --bootstrap ~{bootstrap} \
+    ~{'--outgroup ' + outgroup} \
+    ~{'--date ' + dates_file} \
+    --threads 2
+  >>>
+  output {
+    String date = read_string("DATE")
+    String version = read_string("VERSION")
+    File gubbins_final_tree = "~{cluster_name}.final_tree.tre"
+    File gubbins_final_labelled_tree = "~{cluster_name}.node_labelled.final_tree.tre"
+    File gubbins_polymorphic_fasta = "~{cluster_name}.filtered_polymorphic_sites.fasta"
+    File gubbins_recombination_gff = "~{cluster_name}.recombination_predictions.gff"
+    File gubbins_branch_stats = "~{cluster_name}.per_branch_statistics.csv"
+    File? gubbins_timetree = "~{cluster_name}.final_tree.timetree.tre"
+    File? gubbins_timetree_stats = "~{cluster_name}.lsd.out"
+  }
+  runtime {
+    docker: "~{docker}"
+    memory: "32 GB"
+    cpu: 4
+    disks: "local-disk 100 SSD"
+    preemptible: 0
+    maxRetries: 1
+  }
+}

tasks/phylogenetic_inference/task_iqtree.wdl

@@ -16,7 +16,7 @@ task iqtree {
     iqtree --version | grep version | sed 's/.*version/version/;s/ for Linux.*//' | tee VERSION
 
     numGenomes=`grep -o '>' ~{alignment} | wc -l`
-    if [ $numGenomes -gt 3 ]
+    if [ "$numGenomes" -gt 3 ]
     then
       cp ~{alignment} ./msa.fasta
       iqtree \
@@ -27,13 +27,13 @@ task iqtree {
       -alrt ~{alrt} \
       ~{iqtree_opts}
 
-      cp msa.fasta.contree ~{cluster_name}_msa.tree
+      cp msa.fasta.contree ~{cluster_name}_iqtree.tree
     fi
   >>>
   output {
     String date = read_string("DATE")
     String version = read_string("VERSION")
-    File ml_tree = "~{cluster_name}_msa.tree"
+    File ml_tree = "~{cluster_name}_iqtree.tree"
   }
   runtime {
     docker: "~{docker}"

tasks/phylogenetic_inference/task_snippy_core.wdl

@@ -0,0 +1,56 @@
+version 1.0
+
+task snippy_core {
+  input {
+    Array[File] snippy_variants_outdir_tarball
+    Array[String] samplenames
+    String tree_name
+    String docker = "staphb/snippy:4.6.0"
+    File reference
+    File? bed_file
+  }
+  command <<<
+   # version control
+   snippy --version | head -1 | tee VERSION
+
+   tarball_array=(~{sep=" " snippy_variants_outdir_tarball})
+   samplename_array=(~{sep=" " samplenames})
+
+   # iteratively untar
+   for i in ${tarball_array[@]}; do tar -xf $i; done
+
+   # run snippy core
+   snippy-core \
+   --prefix ~{tree_name} \
+   ~{'--mask ' + bed_file} \
+   --ref ~{reference} \
+   "${samplename_array[@]}"
+
+   # run snippy clean
+   snippy-clean_full_aln \
+   ~{tree_name}.full.aln > ~{tree_name}_snippy_clean_full.aln
+
+   mv ~{tree_name}.aln ~{tree_name}_core.aln
+   mv ~{tree_name}.full.aln ~{tree_name}_full.aln
+   mv ~{tree_name}.tab ~{tree_name}_all_snps.tsv
+   mv ~{tree_name}.txt ~{tree_name}_snps_summary.txt
+  >>>
+  output {
+   String snippy_version = read_string("VERSION")
+   File snippy_core_alignment = "~{tree_name}_core.aln"
+   File snippy_full_alignment = "~{tree_name}_full.aln"
+   File snippy_full_alignment_clean = "~{tree_name}_snippy_clean_full.aln"
+   File snippy_ref = "~{tree_name}.ref.fa"
+   File snippy_core_tab = "~{tree_name}_all_snps.tsv"
+   File snippy_txt = "~{tree_name}_snps_summary.txt"
+   File snippy_vcf = "~{tree_name}.vcf"
+   String snippy_docker_image = docker
+  }
+  runtime {
+    docker: "~{docker}"
+    memory: "8 GB"
+    cpu: 4
+    disks: "local-disk 100 SSD"
+    preemptible: 0
+  }
+}

tasks/phylogenetic_inference/task_snp_dists.wdl

@@ -94,13 +94,15 @@ task snp_dists {
     z.close()
     print "Matrix has been created in current directory as '~{cluster_name}_snp_distance_matrix.tsv.'"
 
-    CODE
+  CODE
+  cp snp-dists-molten-ordered.tsv ~{cluster_name}_snp-dists_list.tsv
+  cp snp-dists-matrix.tsv ~{cluster_name}_snp_distance_matrix.tsv
   >>>
   output {
     String date = read_string("DATE")
     String version = read_string("VERSION")
     File snp_matrix = "${cluster_name}_snp_distance_matrix.tsv"
-    File snp_dists_molten_ordered = "snp-dists-molten-ordered.tsv"
+    File snp_dists_molten_ordered = "${cluster_name}_snp-dists_list.tsv"
   }
   runtime {
     docker: "quay.io/staphb/snp-dists:0.8.2"