Merge pull request #19 from ikmb/devel

marchoeppner · web-flow · commit 5030939e75be · 2023-07-21T08:38:57.000+02:00
Devel
diff --git a/bin/samplesheet_from_folder.rb b/bin/samplesheet_from_folder.rb
@@ -36,6 +36,7 @@
 opts.on("-p", "--[no-]pacbio [FLAG]", TrueClass, "Pacbio data") {|argument| options.pacbio = argument.nil? ? true : argument }
 opts.on("-c","--centre", "=CENTRE","Name of sequencing centre") {|argument| options.centre = argument }
 opts.on("-s","--platform","=PLATFORM", "Name of the sequencing platform") {|argument| options.platform = argument }
+opts.on("-l","--lookup", "=LOOKUP", "Lookup file with lib id <> other id") {|argument| options.lookup = argument }
 opts.on("-h","--help","Display the usage information") {
  puts opts
  exit
@@ -49,12 +50,19 @@
 options.centre ? center = options.centre : center = "IKMB"
 options.platform ? platform = options.platform : platform = "NovaSeq6000"
 
+lookup = {}
+if options.lookup
+	IO.readlines(options.lookup).each do |line|
+		key,value = line.strip.split("\t")
+		lookup[key] = value
+	end
+end
 
 if options.pacbio
 
 	fastq_files = Dir["#{options.folder}/*.fastq.gz"]
 
-	puts "IndivID;SampleID;R1"
+	puts "patient;sample;R1"
 
 	fastq_files.each do |file|
 
@@ -72,27 +80,33 @@
 
 	groups = fastq_files.group_by{|f| f.split("/")[-1].split(/_R[1,2]/)[0] }
 
-	puts "IndivID;SampleID;libraryID;rgID;rgPU;R1;R2"
+	puts "patient;sample;library;rgid;rgpu;R1;R2"
 
 	groups.each do |group, files|
 
-        	left,right = files.sort.collect{|f| File.absolute_path(f)}
+            left,right = files.sort.collect{|f| File.absolute_path(f)}
+
+            abort "Missing one member of the pair for #{group}" unless left && right
+
+            library = group.split("_")[1]
+            sample = library
 
-	        library = group.split("_L00")[0]
-        	sample = group.split("_L00")[0]
+            if lookup.has_key?(library)
+                sample = "#{lookup[library]}"
+            end
 
-	        e = `zcat #{left} | head -n1 `
-		e.gsub!("@", "")
-		header = e
+            e = `zcat #{left} | head -n1 `
+            e.gsub!("@", "")
+            header = e
 
-        	instrument,run_id,flowcell_id,lane,tile,x,y = header.split(" ")[0].split(":")
+            instrument,run_id,flowcell_id,lane,tile,x,y = header.split(" ")[0].split(":")
 
-		index = header.split(" ")[-1].split(":")[-1]
-        	readgroup = flowcell_id + "." + lane + "." + library 
+            index = header.split(" ")[-1].split(":")[-1]
+            readgroup = flowcell_id + "." + lane + "." + library 
 
-	        pgu = flowcell_id + "." + lane + "." + index
+            pgu = flowcell_id + "." + lane + "." + index
 
-        	puts "#{sample};Sample_#{sample};#{library};#{readgroup};#{pgu};#{left};#{right}"
+            puts "#{sample};#{sample};#{library};#{readgroup};#{pgu};#{left};#{right}"
 
 	end
 
diff --git a/modules/helper/bam_select_reads.nf b/modules/helper/bam_select_reads.nf
@@ -18,8 +18,8 @@ process BAM_SELECT_READS {
     path("versions.yml"), emit: versions
 
     script:
-    R1 = bam.getBaseName() + "_R1_001.fastq.gz"
-    R2 = bam.getBaseName() + "_R2_001.fastq.gz"
+    R1 = meta.sample_id + "_R1_001.fastq.gz"
+    R2 = meta.sample_id + "_R2_001.fastq.gz"
 
     """
     samtools view --reference $fasta -hb -o mapped.cram -L $bed $bam
diff --git a/modules/picard/collect_wgs_metrics.nf b/modules/picard/collect_wgs_metrics.nf
@@ -3,7 +3,7 @@ process PICARD_WGS_METRICS {
 
     tag "${meta.patient_id}|${meta.sample_id}"
 
-    publishDir "${params.outdir}/${meta.patient_id}/${meta.sample_id}", mode: 'copy'
+    publishDir "${params.outdir}/${meta.patient_id}/${meta.sample_id}/QC", mode: 'copy'
 
     container "docker://quay.io/biocontainers/picard:2.26.11--hdfd78af_0"
 
diff --git a/subworkflows/trim_and_align.nf b/subworkflows/trim_and_align.nf
@@ -35,9 +35,8 @@ workflow TRIM_AND_ALIGN {
             new_meta = [:]
             new_meta.patient_id = meta.patient_id
             new_meta.sample_id = meta.sample_id
-            def groupKey = meta.sample_id
-            tuple( groupKey, new_meta, bam)
-        }.groupTuple(by: [0,1]).map { g ,new_meta ,bam -> [ new_meta, bam ] }
+            tuple( new_meta, bam)
+        }.groupTuple()
             
         bam_mapped.branch {
             single:   it[1].size() == 1
@@ -60,7 +59,7 @@ workflow TRIM_AND_ALIGN {
 			SAMTOOLS_INDEX.out.bam,
 			ch_fasta
 		)
-
+		
 		ch_versions = ch_versions.mix(SAMTOOLS_MARKDUP.out.versions)
 
 	emit:
diff --git a/workflows/deepvariant_pipeline.nf b/workflows/deepvariant_pipeline.nf
@@ -110,7 +110,7 @@ workflow DEEPVARIANT_PIPELINE {
         if ('intersect' in tools) {
             BAM_SELECT_READS(
                 TRIM_AND_ALIGN.out.bam,
-                ch_bed,
+                ch_bed.collect(),
                 ch_fasta
             )
 
@@ -121,7 +121,7 @@ workflow DEEPVARIANT_PIPELINE {
         if ('deepvariant' in tools) {
             DEEPVARIANT_SHORT_READS(
                 TRIM_AND_ALIGN.out.bam,
-                ch_bed,
+                ch_bed.collect(),
                 ch_fasta
             )
             ch_vcf = ch_vcf.mix(DEEPVARIANT_SHORT_READS.out.vcf)
@@ -150,13 +150,13 @@ workflow DEEPVARIANT_PIPELINE {
     MOSDEPTH(
         ch_bam,
         ch_fasta,
-        ch_bed
+        ch_bed.collect()
     )
     
     PICARD_WGS_METRICS(
         ch_bam,
         ch_fasta,
-        ch_bed
+        ch_bed.collect()
     )
 
     CUSTOM_DUMPSOFTWAREVERSIONS (
@@ -176,16 +176,14 @@ workflow DEEPVARIANT_PIPELINE {
 
 def create_fastq_channel(LinkedHashMap row) {
 
-    // IndivID;SampleID;libraryID;rgID;rgPU;platform;platform_model;Center;Date;R1;R2
+    // patient;sample;library;rgid;rgpu;R1;R2
 
     def meta = [:]
-    meta.patient_id = row.IndivID
-    meta.sample_id = row.SampleID
-    meta.library_id = row.libraryID
-    meta.readgroup_id = row.rgID
-    meta.center = row.Center
-    meta.date = row.Date
-    meta.platform_unit = row.rgPU
+    meta.patient_id = row.patient
+    meta.sample_id = row.sample
+    meta.library_id = row.library
+    meta.readgroup_id = row.rgid
+    meta.platform_unit = row.rgpu
 
     def array = []
     array = [ meta, file(row.R1), file(row.R2) ]
@@ -195,11 +193,11 @@ def create_fastq_channel(LinkedHashMap row) {
 
 def create_pacbio_channel(LinkedHashMap row) {
 
-    // IndivID;SampleID;libraryID;rgID;rgPU;platform;platform_model;Center;Date;R1;R2
+    // patient;sample;R1
 
     def meta = [:]
-    meta.patient_id = row.IndivID
-    meta.sample_id = row.SampleID
+    meta.patient_id = row.patient
+    meta.sample_id = row.sample
 
     def array = []
     array = [ meta, file(row.R1) ]
