Merge pull request #115 from nf-core/support-donor-sample

lib/Constants.groovy

@@ -93,6 +93,7 @@ class Constants {
         TUMOR,
         NORMAL,
         TUMOR_NORMAL,
+        DONOR,
     }
 
     static enum SequenceType {
@@ -136,6 +137,18 @@ class Constants {
             SequenceType.DNA,
         ],
 
+        BAM_DNA_DONOR: [
+            FileType.BAM,
+            SampleType.DONOR,
+            SequenceType.DNA,
+        ],
+
+        BAM_MARKDUPS_DNA_DONOR: [
+            FileType.BAM_MARKDUPS,
+            SampleType.DONOR,
+            SequenceType.DNA,
+        ],
+
         BAM_RNA_TUMOR: [
             FileType.BAM,
             SampleType.TUMOR,
@@ -154,6 +167,12 @@ class Constants {
             SequenceType.DNA,
         ],
 
+        BAI_DNA_DONOR: [
+            FileType.BAI,
+            SampleType.DONOR,
+            SequenceType.DNA,
+        ],
+
         BAI_RNA_TUMOR: [
             FileType.BAI,
             SampleType.TUMOR,

lib/Utils.groovy

@@ -267,6 +267,13 @@ class Utils {
 
             }
 
+            // Do not allow donor sample without normal sample
+            if (Utils.hasDonorDna(meta) && ! Utils.hasNormalDna(meta)) {
+                log.error "a donor sample but not normal sample was found for ${meta.group_id}\n\n" +
+                    "Analysis with a donor sample requires a normal sample."
+                Nextflow.exit(1)
+            }
+
             // Apply some required restrictions to targeted mode
             if (run_config.mode === Constants.RunMode.TARGETED) {
 
@@ -401,6 +408,9 @@ class Utils {
         return meta.getOrDefault([Constants.SampleType.NORMAL, Constants.SequenceType.DNA], [:])
     }
 
+    static public getDonorDnaSample(meta) {
+        return meta.getOrDefault([Constants.SampleType.DONOR, Constants.SequenceType.DNA], [:])
+    }
 
     // Sample names
     static public getTumorDnaSampleName(meta) {
@@ -415,8 +425,12 @@ class Utils {
         return getNormalDnaSample(meta)['sample_id']
     }
 
+    static public getDonorDnaSampleName(meta) {
+        return getDonorDnaSample(meta)['sample_id']
+    }
+
 
-    // Files
+    // Files - Tumor DNA
     static public getTumorDnaFastq(meta) {
         return getTumorDnaSample(meta).getOrDefault(Constants.FileType.FASTQ, null)
     }
@@ -447,6 +461,7 @@ class Utils {
     }
 
 
+    // Files - Normal DNA
     static public getNormalDnaFastq(meta) {
         return getNormalDnaSample(meta).getOrDefault(Constants.FileType.FASTQ, null)
     }
@@ -475,7 +490,6 @@ class Utils {
         return getNormalDnaMarkdupsBam(meta) !== null
     }
 
-
     static public hasDnaFastq(meta) {
         return hasNormalDnaFastq(meta) || hasTumorDnaFastq(meta)
     }
@@ -485,6 +499,38 @@ class Utils {
     }
 
 
+    // Files - Donor DNA
+    static public getDonorDnaFastq(meta) {
+        return getDonorDnaSample(meta).getOrDefault(Constants.FileType.FASTQ, null)
+    }
+
+    static public getDonorDnaBam(meta) {
+        return getDonorDnaSample(meta).getOrDefault(Constants.FileType.BAM, null)
+    }
+
+    static public getDonorDnaMarkdupsBam(meta) {
+        return getDonorDnaSample(meta).getOrDefault(Constants.FileType.BAM_MARKDUPS, null)
+    }
+
+    static public getDonorDnaBai(meta) {
+        return getDonorDnaSample(meta).getOrDefault(Constants.FileType.BAI, null)
+    }
+
+
+    static public hasDonorDnaFastq(meta) {
+        return getDonorDnaFastq(meta) !== null
+    }
+
+    static public hasDonorDnaBam(meta) {
+        return getDonorDnaBam(meta) !== null
+    }
+
+    static public hasDonorDnaMarkdupsBam(meta) {
+        return getDonorDnaMarkdupsBam(meta) !== null
+    }
+
+
+    // Files - Tumor RNA
     static public getTumorRnaFastq(meta) {
         return getTumorRnaSample(meta).getOrDefault(Constants.FileType.FASTQ, null)
     }
@@ -516,6 +562,10 @@ class Utils {
         return hasNormalDnaBam(meta) || hasNormalDnaMarkdupsBam(meta) || hasNormalDnaFastq(meta)
     }
 
+    static public hasDonorDna(meta) {
+        return hasDonorDnaBam(meta) || hasDonorDnaMarkdupsBam(meta) || hasDonorDnaFastq(meta)
+    }
+
     static public hasTumorRna(meta) {
         return hasTumorRnaBam(meta) || hasTumorRnaFastq(meta)
     }

modules/local/amber/main.nf

@@ -8,7 +8,7 @@ process AMBER {
         'biocontainers/hmftools-amber:4.0.1--hdfd78af_0' }"
 
     input:
-    tuple val(meta), path(tumor_bam), path(normal_bam), path(tumor_bai), path(normal_bai)
+    tuple val(meta), path(tumor_bam), path(normal_bam), path(donor_bam), path(tumor_bai), path(normal_bai), path(donor_bai)
     val genome_ver
     path heterozygous_sites
     path target_region_bed
@@ -25,10 +25,17 @@ process AMBER {
 
     def xmx_mod = task.ext.xmx_mod ?: 0.75
 
-    def reference_arg = meta.containsKey('normal_id') ? "-reference ${meta.normal_id}" : ''
-    def reference_bam_arg = normal_bam ? "-reference_bam ${normal_bam}" : ''
+    def reference_ids = []
+    if (meta.normal_id != null) reference_ids.add(meta.normal_id)
+    if (meta.donor_id != null) reference_ids.add(meta.donor_id)
+    def reference_arg = reference_ids.size() > 0 ? "-reference ${String.join(",", reference_ids)}" : ""
 
-    def target_regions_bed_arg = target_region_bed ? "-target_regions_bed ${target_region_bed}" : ''
+    def reference_bams = []
+    if (normal_bam) reference_bams.add(normal_bam.toString())
+    if (donor_bam) reference_bams.add(donor_bam.toString())
+    def reference_bam_arg = reference_bams.size() > 0 ? "-reference_bam ${String.join(",", reference_bams)}" : ""
+
+    def target_regions_bed_arg = target_region_bed ? "-target_regions_bed ${target_region_bed}" : ""
 
     """
     amber \\

modules/local/amber/meta.yml

@@ -23,6 +23,10 @@ input:
       type: file
       description: Normal BAM file
       pattern: "*.{bam}"
+  - donor_bam:
+      type: file
+      description: Donor BAM file
+      pattern: "*.{bam}"
   - tumor_bai:
       type: file
       description: Tumor BAI file
@@ -31,6 +35,10 @@ input:
       type: file
       description: Normal BAI file
       pattern: "*.{bai}"
+  - donor_bai:
+      type: file
+      description: Donor BAI file
+      pattern: "*.{bai}"
   - genome_ver:
       type: string
       description: Reference genome version

modules/local/sage/somatic/main.nf

@@ -10,7 +10,7 @@ process SAGE_SOMATIC {
         'biocontainers/hmftools-sage:3.4.4--hdfd78af_0' }"
 
     input:
-    tuple val(meta), path(tumor_bam), path(normal_bam), path(tumor_bai), path(normal_bai)
+    tuple val(meta), path(tumor_bam), path(normal_bam), path(donor_bam), path(tumor_bai), path(normal_bai), path(donor_bai)
     path genome_fasta
     val genome_ver
     path genome_fai
@@ -34,8 +34,17 @@ process SAGE_SOMATIC {
 
     def xmx_mod = task.ext.xmx_mod ?: 0.95
 
-    def reference_arg = meta.containsKey('normal_id') ? "-reference ${meta.normal_id}" : ''
-    def reference_bam_arg = normal_bam ? "-reference_bam ${normal_bam}" : ''
+    def reference_ids = []
+    if (meta.normal_id != null) reference_ids.add(meta.normal_id)
+    if (meta.donor_id != null) reference_ids.add(meta.donor_id)
+    def reference_arg = reference_ids.size() > 0 ? "-reference ${String.join(",", reference_ids)}" : ""
+
+    def reference_bams = []
+    if (normal_bam) reference_bams.add(normal_bam.toString())
+    if (donor_bam) reference_bams.add(donor_bam.toString())
+    def reference_bam_arg = reference_bams.size() > 0 ? "-reference_bam ${String.join(",", reference_bams)}" : ""
+
+    def ref_sample_count_arg = "-ref_sample_count ${reference_ids.size()}"
 
     """
     mkdir -p somatic/
@@ -45,6 +54,7 @@ process SAGE_SOMATIC {
         ${args} \\
         ${reference_arg} \\
         ${reference_bam_arg} \\
+        ${ref_sample_count_arg} \\
         -tumor ${meta.tumor_id} \\
         -tumor_bam ${tumor_bam} \\
         -ref_genome ${genome_fasta} \\

modules/local/sage/somatic/meta.yml

@@ -25,6 +25,10 @@ input:
       type: file
       description: Normal BAM file
       pattern: "*.{bam}"
+  - donor_bam:
+      type: file
+      description: Donor BAM file
+      pattern: "*.{bam}"
   - tumor_bai:
       type: file
       description: Tumor BAI file
@@ -33,6 +37,10 @@ input:
       type: file
       description: Normal BAI file
       pattern: "*.{bai}"
+  - donor_bai:
+      type: file
+      description: Donor BAI file
+      pattern: "*.{bai}"
   - genome_fasta:
       type: file
       description: Reference genome assembly FASTA file

subworkflows/local/amber_profiling/main.nf

@@ -13,6 +13,7 @@ workflow AMBER_PROFILING {
     ch_inputs          // channel: [mandatory] [ meta ]
     ch_tumor_bam       // channel: [mandatory] [ meta, bam, bai ]
     ch_normal_bam      // channel: [mandatory] [ meta, bam, bai ]
+    ch_donor_bam       // channel: [mandatory] [ meta, bam, bai ]
 
     // Reference data
     genome_version     // channel: [mandatory] genome version
@@ -30,27 +31,32 @@ workflow AMBER_PROFILING {
     ch_inputs_sorted = WorkflowOncoanalyser.groupByMeta(
         ch_tumor_bam,
         ch_normal_bam,
+        ch_donor_bam,
     )
-        .map { meta, tumor_bam, tumor_bai, normal_bam, normal_bai ->
+        .map { meta, tumor_bam, tumor_bai, normal_bam, normal_bai, donor_bam, donor_bai ->
             return [
                 meta,
                 Utils.selectCurrentOrExisting(tumor_bam, meta, Constants.INPUT.BAM_MARKDUPS_DNA_TUMOR),
                 tumor_bai ?: Utils.getInput(meta, Constants.INPUT.BAI_DNA_TUMOR),
+
                 Utils.selectCurrentOrExisting(normal_bam, meta, Constants.INPUT.BAM_MARKDUPS_DNA_NORMAL),
                 normal_bai ?: Utils.getInput(meta, Constants.INPUT.BAI_DNA_NORMAL),
+
+                Utils.selectCurrentOrExisting(donor_bam, meta, Constants.INPUT.BAM_MARKDUPS_DNA_DONOR),
+                donor_bai ?: Utils.getInput(meta, Constants.INPUT.BAI_DNA_DONOR),
             ]
         }
-        .branch { meta, tumor_bam, tumor_bai, normal_bam, normal_bai ->
+        .branch { meta, tumor_bam, tumor_bai, normal_bam, normal_bai, donor_bam, donor_bai ->
             def has_existing = Utils.hasExistingInput(meta, Constants.INPUT.AMBER_DIR)
             runnable: tumor_bam && !has_existing
             skip: true
                 return meta
         }
 
     // Create process input channel
-    // channel: [ meta_amber, tumor_bam, normal_bam, tumor_bai, normal_bai ]
+    // channel: [ meta_amber, tumor_bam, normal_bam, donor_bam, tumor_bai, normal_bai, donor_bai ]
     ch_amber_inputs = ch_inputs_sorted.runnable
-        .map { meta, tumor_bam, tumor_bai, normal_bam, normal_bai ->
+        .map { meta, tumor_bam, tumor_bai, normal_bam, normal_bai, donor_bam, donor_bai ->
 
             def meta_amber = [
                 key: meta.group_id,
@@ -62,7 +68,11 @@ workflow AMBER_PROFILING {
                 meta_amber.normal_id = Utils.getNormalDnaSampleName(meta)
             }
 
-            [meta_amber, tumor_bam, normal_bam, tumor_bai, normal_bai]
+            if (donor_bam) {
+                meta_amber.donor_id = Utils.getDonorDnaSampleName(meta)
+            }
+
+            [meta_amber, tumor_bam, normal_bam, donor_bam, tumor_bai, normal_bai, donor_bai]
         }
 
     // Run process

subworkflows/local/read_alignment_dna/main.nf

@@ -41,12 +41,20 @@ workflow READ_ALIGNMENT_DNA {
             skip: true
         }
 
+    ch_inputs_donor_sorted = ch_inputs
+        .branch { meta ->
+            def has_existing = Utils.hasExistingInput(meta, Constants.INPUT.BAM_DNA_DONOR)
+            runnable: Utils.hasDonorDnaFastq(meta) && !has_existing
+            skip: true
+        }
+
     // Create FASTQ input channel
     // channel: [ meta_fastq, fastq_fwd, fastq_rev ]
     ch_fastq_inputs = Channel.empty()
         .mix(
             ch_inputs_tumor_sorted.runnable.map { meta -> [meta, Utils.getTumorDnaSample(meta), 'tumor'] },
             ch_inputs_normal_sorted.runnable.map { meta -> [meta, Utils.getNormalDnaSample(meta), 'normal'] },
+            ch_inputs_donor_sorted.runnable.map { meta -> [meta, Utils.getDonorDnaSample(meta), 'donor'] },
         )
         .flatMap { meta, meta_sample, sample_type ->
             meta_sample
@@ -185,9 +193,10 @@ workflow READ_ALIGNMENT_DNA {
         }
         .groupTuple()
         .branch { meta_group, bams, bais ->
-            assert ['tumor', 'normal'].contains(meta_group.sample_type)
+            assert ['tumor', 'normal', 'donor'].contains(meta_group.sample_type)
             tumor: meta_group.sample_type == 'tumor'
             normal: meta_group.sample_type == 'normal'
+            donor: meta_group.sample_type == 'donor'
             placeholder: true
         }
 
@@ -205,9 +214,16 @@ workflow READ_ALIGNMENT_DNA {
             ch_inputs_normal_sorted.skip.map { meta -> [meta, [], []] },
         )
 
+    ch_bam_donor_out = Channel.empty()
+        .mix(
+            WorkflowOncoanalyser.restoreMeta(ch_bams_united.donor, ch_inputs),
+            ch_inputs_donor_sorted.skip.map { meta -> [meta, [], []] },
+        )
+
     emit:
     dna_tumor  = ch_bam_tumor_out  // channel: [ meta, [bam, ...], [bai, ...] ]
     dna_normal = ch_bam_normal_out // channel: [ meta, [bam, ...], [bai, ...] ]
+    dna_donor  = ch_bam_donor_out  // channel: [ meta, [bam, ...], [bai, ...] ]
 
     versions   = ch_versions       // channel: [ versions.yml ]
 }