Add Arriba (and respective pVACfuse step) to the immuno workflow

definitions/immuno.wdl

@@ -37,6 +37,7 @@ struct Rnaseq {
   Array[File] stringtie_expression
   Array[File] kallisto_expression
   StarFusion star_fusion
+  Array[File] arriba
   Array[File] fusioninspector_evidence
 }
 
@@ -501,7 +502,7 @@ workflow immuno {
     anchor_contribution_threshold=anchor_contribution_threshold
   }
 
-  call pf.pvacfuse {
+  call pf.pvacfuse as agfusion_pvacfuse {
     input:
     input_fusions_zip=rna.annotated_fusion_predictions_zip,
     star_fusion_file=rna.star_fusion_abridge,
@@ -532,6 +533,37 @@ workflow immuno {
     problematic_amino_acids=problematic_amino_acids,
   }
 
+  call pf.pvacfuse as arriba_pvacfuse {
+    input:
+    input_fusions_zip=rna.arriba_fusion_predict,
+    star_fusion_file=rna.star_fusion_abridge,
+    sample_name=tumor_sample_name,
+    alleles=hlaConsensus.consensus_alleles,
+    prediction_algorithms=prediction_algorithms,
+    epitope_lengths_class_i=epitope_lengths_class_i,
+    epitope_lengths_class_ii=epitope_lengths_class_ii,
+    binding_threshold=binding_threshold,
+    percentile_threshold=percentile_threshold,
+    iedb_retries=iedb_retries,
+    keep_tmp_files=pvacfuse_keep_tmp_files,
+    net_chop_method=net_chop_method,
+    netmhc_stab=netmhc_stab,
+    top_score_metric=top_score_metric,
+    net_chop_threshold=net_chop_threshold,
+    run_reference_proteome_similarity=run_reference_proteome_similarity,
+    peptide_fasta=peptide_fasta,
+    additional_report_columns=additional_report_columns,
+    fasta_size=fasta_size,
+    downstream_sequence_length=downstream_sequence_length,
+    exclude_nas=exclude_nas,
+    n_threads=pvacseq_threads,
+    read_support=pvacfuse_read_support,
+    expn_val=pvacfuse_expn_val,
+    allele_specific_binding_thresholds=allele_specific_binding_thresholds,
+    aggregate_inclusion_binding_threshold=aggregate_inclusion_binding_threshold,
+    problematic_amino_acids=problematic_amino_acids,
+  }
+
   call generate_fda_metrics.generateFdaMetrics {
     input:
       reference = reference,
@@ -606,6 +638,10 @@ workflow immuno {
         ],
         candidates_preliminary: rna.prelim_starfusion_results
       },
+      arriba: [
+        rna.arriba_fusion_predict,
+        rna.arriba_fusion_discard
+      ],
       fusioninspector_evidence: rna.fusioninspector_evidence
     }
 
@@ -732,10 +768,16 @@ workflow immuno {
       phase_vcf: [phaseVcf.phased_vcf, phaseVcf.phased_vcf_tbi]
     }
 
-    MHC pVACfuse = object {
-      mhc_i: pvacfuse.mhc_i,
-      mhc_ii: pvacfuse.mhc_ii,
-      combined: pvacfuse.combined
+    MHC agfusion_pvacfuse_predictions = object {
+      mhc_i: agfusion_pvacfuse.mhc_i,
+      mhc_ii: agfusion_pvacfuse.mhc_ii,
+      combined: agfusion_pvacfuse.combined
+    }
+
+    MHC arriba_pvacfuse_predictions = object {
+      mhc_i: arriba_pvacfuse.mhc_i,
+      mhc_ii: arriba_pvacfuse.mhc_ii,
+      combined: arriba_pvacfuse.combined
     }
 
     File pvacseq_annotated_expression_vcf_gz = pvacseq.annotated_vcf

definitions/rnaseq_star_fusion.wdl

@@ -10,6 +10,7 @@ import "tools/kallisto.wdl" as k
 import "tools/mark_duplicates_and_sort.wdl" as mdas
 import "tools/samtools_sort.wdl" as ss
 import "tools/star_fusion_detect.wdl" as sfd
+import "tools/arriba.wdl" as ar
 import "tools/strandedness_check.wdl" as sc
 import "tools/stringtie.wdl" as s
 import "tools/transcript_to_gene.wdl" as ttg
@@ -84,6 +85,16 @@ workflow rnaseqStarFusion {
     min_ffpm_level=min_ffpm_level
   }
 
+  call ar.arriba {
+    input:
+    reference_annotation=reference_annotation,
+    reference=reference,
+    reference_fai=reference_fai,
+    reference_dict=reference_dict,
+    aligned_bam=starFusionDetect.aligned_bam,
+    star_fusion_chim_junc=starFusionDetect.chim_junc
+  }
+
   call k.kallisto {
     input:
     kallisto_index=kallisto_index,
@@ -173,6 +184,8 @@ workflow rnaseqStarFusion {
     File final_bam_bai = indexBam.indexed_bam_bai
     File final_bai = indexBam.indexed_bai
     File annotated_fusion_predictions_zip = agfusion.annotated_fusion_predictions_zip
+    File arriba_fusion_predict=arriba.fusion_predictions
+    File arriba_fusion_discard=arriba.discarded_fusion_predictions
     File? star_fusion_coding_region_effects = starFusionDetect.coding_region_effects
     Array[File] fusioninspector_evidence = starFusionDetect.fusioninspector_evidence
     File flagstats = samtoolsFlagstat.flagstats

definitions/tools/arriba.wdl

@@ -0,0 +1,46 @@
+version 1.0
+
+task arriba {
+  input {
+    File aligned_bam
+    File reference_annotation
+    File reference
+    File reference_fai
+    File reference_dict
+    File star_fusion_chim_junc
+  }
+
+  Float bam_size_gb = size([aligned_bam, reference_annotation], "GB")
+  Float reference_size_gb = size([reference, reference_fai, reference_dict], "GB")
+  Int space_needed_gb = 10 + round(3*bam_size_gb + reference_size_gb)
+
+  runtime {
+    preemptible: 1
+    maxRetries: 2
+    memory: "64GB"
+    docker: "uhrigs/arriba:2.4.0"
+    disks: "local-disk ~{space_needed_gb} HDD"
+  }
+
+  # explicit typing required, don't inline
+  command <<<
+    /arriba_v2.4.0/arriba \
+    -b /arriba_v2.4.0/database/blacklist_hg38_GRCh38_v2.4.0.tsv.gz \
+    -k /arriba_v2.4.0/database/known_fusions_hg38_GRCh38_v2.4.0.tsv.gz \
+    -t /arriba_v2.4.0/database/known_fusions_hg38_GRCh38_v2.4.0.tsv.gz \
+    -p /arriba_v2.4.0/database/protein_domains_hg38_GRCh38_v2.4.0.gff3 \
+    -o arriba_fusions.tsv \
+    -O arriba_fusions.discarded.tsv \
+    -x ~{aligned_bam} \
+    -g ~{reference_annotation} \
+    -a ~{reference} \
+    -c ~{star_fusion_chim_junc}
+  >>>
+
+  output {
+    File fusion_predictions = "arriba_fusions.tsv"
+    File discarded_fusion_predictions = "arriba_fusions.discarded.tsv"
+  }
+}
+
+workflow wf { call arriba { input: } }

definitions/tools/pvacfuse.wdl

@@ -35,7 +35,7 @@ task pvacfuse {
   runtime {
     preemptible: 1
     maxRetries: 2
-    docker: "susannakiwala/pvactools:4.0.0_rc_pvacview_v13"
+    docker: "susannakiwala/pvactools:4.0.0_rc_pvacview_v14"
     memory: "16GB"
     cpu: n_threads
     disks: "local-disk ~{space_needed_gb} HDD"

definitions/tools/pvacseq.wdl

@@ -56,7 +56,7 @@ task pvacseq {
     maxRetries: 2
     memory: "16GB"
     cpu: n_threads
-    docker: "susannakiwala/pvactools:4.0.0_rc_pvacview_v13"
+    docker: "susannakiwala/pvactools:4.0.0_rc_pvacview_v14"
     disks: "local-disk ~{space_needed_gb} HDD"
   }
 

definitions/tools/star_fusion_detect.wdl

@@ -48,7 +48,8 @@ task starFusionDetect {
         ~{if defined(fusioninspector_mode) then "--FusionInspector " + fusioninspector_mode else ""} \
         --STAR_outSAMattrRGline "~{sep=" , " outsam_attrrg_line}" \
         --left_fq ~{sep="," fastq} --right_fq ~{sep="," fastq2} \
-        --min_FFPM ~{min_ffpm_level}
+        --min_FFPM ~{min_ffpm_level} \
+        --chimOutType SeparateSAMold
   >>>
 
   output {
@@ -66,7 +67,7 @@ task starFusionDetect {
     File log = fusion_output_dir + "/Log.out"
     File log_progress = fusion_output_dir + "/Log.progress.out"
     File splice_junction_out = fusion_output_dir + "/SJ.out.tab"
-    File chim_junc = fusion_output_dir + "/Chimeric.out.junction"
+    File chim_junc = fusion_output_dir + "/Chimeric.out.sam"
     # STAR also outputs gene counts file just like Kallisto
     File gene_counts = fusion_output_dir + "/ReadsPerGene.out.tab"