Apollo_Illumina_PE & kSNP3 workflows (#7)

* Update task_taxon_id.wdl

* refine output tsv

* increase compure resources

* fix formatting

* parse module reports to grab relevant outputs

* change local dev workflow name

* write to predicted taxon; format delta to two decimal places

* fix syntax

* print info for failed species predictions

* print info for failed species predictions

* print info for failed species predictions

* fix syntax

* update no-call message

* split taxon to genus and species

* write no-calls out as string

* fix typo

* rename to apollo

* add ksnp3 workflow

* Switch to gambit naming

* Fix file path

* fix typo

* fix typo

.dockstore.yml

@@ -5,3 +5,13 @@ workflows:
    primaryDescriptorPath: /workflows/wf_apollo_pe.wdl
    testParameterFiles:
     - empty.json
+ - name: kSNP3
+   subclass: WDL
+   primaryDescriptorPath: /workflows/wf_ksnp3.wdl
+   testParameterFiles:
+    - empty.json
+ - name: Apollo_Illumina_PE
+   subclass: WDL
+   primaryDescriptorPath: /workflows/wf_apollo_illumina_pe.wdl
+   testParameterFiles:
+    - empty.json

tasks/task_phylo.wdl

@@ -0,0 +1,91 @@
+version 1.0
+
+task ksnp3 {
+
+  input {
+    Array[File] assembly_fasta
+    Array[String] samplename
+    String cluster_name
+    Int  kmer_size  =  19
+    String    docker_image = "staphb/ksnp3:3.1"
+    Int       mem_size_gb = 8
+    Int       CPUs = 4
+  }
+
+  command <<<
+
+  assembly_array=(~{sep=' ' assembly_fasta})
+  assembly_array_len=$(echo "${#assembly_array[@]}")
+  samplename_array=(~{sep=' ' samplename})
+  samplename_array_len=$(echo "${#samplename_array[@]}")
+
+  # Ensure assembly, and samplename arrays are of equal length
+  if [ "$assembly_array_len" -ne "$samplename_array_len" ]; then
+    echo "Assembly array (length: $assembly_array_len) and samplename array (length: $samplename_array_len) are of unequal length." >&2
+    exit 1
+  fi
+
+  # create file of filenames for kSNP3 input
+  date 
+
+  touch ksnp3_input.tsv
+  for index in ${!assembly_array[@]}; do
+    assembly=${assembly_array[$index]}
+    samplename=${samplename_array[$index]}
+
+    echo -e "${assembly}\t${samplename}" >> ksnp3_input.tsv
+  done
+
+  kSNP3 -in ksnp3_input.tsv -outdir ksnp3 -k ~{kmer_size} -core 
+
+  mv ksnp3/core_SNPs_matrix.fasta ~{cluster_name}_core_SNPs_matrix.fasta
+  mv ksnp3/tree.core.tre ~{cluster_name}_core.tree
+
+  >>>
+
+  output {
+    File    ksnp3_matrix   = "${cluster_name}_core_SNPs_matrix.fasta"
+    File    ksnp3_tree  = "${cluster_name}_core.tree"
+    String   ksnp3_docker_image = docker_image
+
+  }
+
+  runtime {
+    docker:       docker_image
+    memory:        "~{mem_size_gb} GB"
+    cpu:          CPUs
+    disks:        "local-disk 100 SSD"
+    preemptible:  0
+  }
+}
+
+task snp_dists {
+
+  input {
+    File      alignment
+    String    cluster_name
+  }
+
+  command{
+    # date and version control
+    date | tee DATE
+    snp-dists -v | tee VERSION
+
+    snp-dists ${alignment} > ${cluster_name}_snp_distance_matrix.tsv
+
+}
+
+  output {
+    String     date = read_string("DATE")
+    String     version = read_string("VERSION")
+    File       snp_matrix = "${cluster_name}_snp_distance_matrix.tsv"
+  }
+
+  runtime {
+    docker:       "staphb/snp-dists:0.6.2"
+    memory:       "2 GB"
+    cpu:          2
+    disks:        "local-disk 100 SSD"
+    preemptible:  0
+  }
+}

tasks/task_taxon_id.wdl

@@ -1,30 +1,34 @@
 version 1.0
 
-task midas_nsphl {
+task gambit {
   input {
-    File assembly
-    String samplename
-    String docker="theiagen/midas_nsphl:1.0.0"
+    File    assembly
+    String  samplename
+    String  docker="theiagen/midas_nsphl:1.0.0"
   }
   command <<<
     # capture date and version
     date | tee DATE
 
-    midas query ~{assembly} | tail -n2 > ~{samplename}_midas_nsphl.csv
+    midas query ~{assembly} | tail -n2 > ~{samplename}_gambit.csv
 
     python3 <<CODE
     import csv
     #grab output genome length and number contigs by column header
-    with open("~{samplename}_midas_nsphl.csv",'r') as csv_file:
+    with open("~{samplename}_gambit.csv",'r') as csv_file:
       csv_reader = list(csv.DictReader(csv_file, delimiter=","))
       for line in csv_reader:
-        with open("MIDAS_DELTA", 'wt') as midas_delta:
+        with open ("GAMBIT_SCORE", 'wt') as gambit_score:
+          top_score=float(line["top_score"])
+          top_score="{:.2f}".format(top_score)
+          gambit_score.write(str(top_score))
+        with open("GAMBIT_DELTA", 'wt') as gambit_delta:
           top_score=float(line["top_score"])
           species_threshold=float(line["species_threshold"])
           delta=top_score - species_threshold
           #format delta to two decimal placesn
           delta="{:.2f}".format(delta)
-          midas_delta.write(str(delta))
+          gambit_delta.write(str(delta))
         with open("PREDICTED_GENUS", 'wt') as predicted_genus:
           genus=line["predicted_genus"]
           if not genus:
@@ -38,11 +42,12 @@ task midas_nsphl {
     CODE
   >>>
   output {
-    File midas_nsphl_report = "~{samplename}_midas_nsphl.csv"
-    String  midas_nsphl_docker   = docker
-    String  pipeline_date = read_string("DATE")
-    Float midas_delta = read_float("MIDAS_DELTA")
-    String predicted_genus = read_string("PREDICTED_GENUS")
+    File   gambit_report     = "~{samplename}_gambit.csv"
+    String gambit_docker     = docker
+    String pipeline_date     = read_string("DATE")
+    Float  gambit_score      = read_float("GAMBIT_SCORE") 
+    Float  gambit_delta      = read_float("GAMBIT_DELTA")
+    String predicted_genus   = read_string("PREDICTED_GENUS")
     String predicted_species = read_string("PREDICTED_SPECIES")
   }
   runtime {

workflows/wf_apollo_pe.wdl → workflows/wf_apollo_illumina_pe.wdl

@@ -5,7 +5,7 @@ import "../tasks/task_qc_utils.wdl" as qc
 import "../tasks/task_taxon_id.wdl" as taxon_id
 import "../tasks/task_denovo_assembly.wdl" as assembly
 
-workflow apollo_pe {
+workflow apollo_illumina_pe {
   meta {
     description: "De-novo genome assembly, taxonomic ID, and QC of paired-end bacterial NGS data"
   }
@@ -41,7 +41,7 @@ workflow apollo_pe {
       samplename = samplename,
       genome_length = quast.genome_length
   }
-  call taxon_id.midas_nsphl {
+  call taxon_id.gambit {
     input:
       assembly = shovill_pe.assembly_fasta,
       samplename = samplename
@@ -75,11 +75,12 @@ workflow apollo_pe {
   Float r2_mean_q = cg_pipeline.r2_mean_q
   Float est_coverage = cg_pipeline.est_coverage
 
-  File  midas_nsphl_report = midas_nsphl.midas_nsphl_report
-  String  midas_nsphl_docker = midas_nsphl.midas_nsphl_docker
-  Float midas_delta = midas_nsphl.midas_delta
-  String predicted_genus = midas_nsphl.predicted_genus
-  String predicted_species = midas_nsphl.predicted_species
+  File  gambit_report = gambit.gambit_report
+  String  gambit_docker = gambit.gambit_docker
+  Float gambit_score = gambit.gambit_score
+  Float gambit_delta = gambit.gambit_delta
+  String predicted_genus = gambit.predicted_genus
+  String predicted_species = gambit.predicted_species
 
   }
 }

workflows/wf_ksnp3.wdl

@@ -0,0 +1,28 @@
+version 1.0
+
+import "../tasks/task_phylo.wdl" as phylo
+
+workflow ksnp3 {
+	input {
+		Array[File] assembly_fasta
+    Array[String] samplename
+    String cluster_name
+	}
+	call phylo.ksnp3 as ksnp3_task {
+		input:
+			assembly_fasta=assembly_fasta,
+      samplename=samplename,
+      cluster_name=cluster_name
+	}
+  call phylo.snp_dists {
+    input:
+      cluster_name = cluster_name,
+      alignment = ksnp3_task.ksnp3_matrix
+  }
+
+    output {
+      File    snp_matrix   = snp_dists.snp_matrix
+      File    ksnp3_tree  = ksnp3_task.ksnp3_tree
+      String   ksnp3_docker = ksnp3_task.ksnp3_docker_image
+    }
+}