add cram support to bowtie2 (nf-core#5180)

* add cram support to bowtie2 * fix tests * fastq_align_bowtie2 > nf-test * fix tests * more robust test * fix cram tests
rki-mf1 · Mar 19, 2024 · 0fe3083 · 0fe3083
1 parent 176c11c
commit 0fe3083
Show file tree

Hide file tree

Showing 13 changed files with 507 additions and 266 deletions.
diff --git a/modules/nf-core/bowtie2/align/main.nf b/modules/nf-core/bowtie2/align/main.nf
@@ -10,13 +10,18 @@ process BOWTIE2_ALIGN {
     input:
     tuple val(meta) , path(reads)
     tuple val(meta2), path(index)
+    tuple val(meta3), path(fasta)
     val   save_unaligned
     val   sort_bam
 
     output:
-    tuple val(meta), path("*.{bam,sam}"), emit: aligned
+    tuple val(meta), path("*.sam")      , emit: sam     , optional:true
+    tuple val(meta), path("*.bam")      , emit: bam     , optional:true
+    tuple val(meta), path("*.cram")     , emit: cram    , optional:true
+    tuple val(meta), path("*.csi")      , emit: csi     , optional:true
+    tuple val(meta), path("*.crai")     , emit: crai    , optional:true
     tuple val(meta), path("*.log")      , emit: log
-    tuple val(meta), path("*fastq.gz")  , emit: fastq, optional:true
+    tuple val(meta), path("*fastq.gz")  , emit: fastq   , optional:true
     path  "versions.yml"                , emit: versions
 
     when:
@@ -39,7 +44,10 @@ process BOWTIE2_ALIGN {
 
     def samtools_command = sort_bam ? 'sort' : 'view'
     def extension_pattern = /(--output-fmt|-O)+\s+(\S+)/
-    def extension = (args2 ==~ extension_pattern) ? (args2 =~ extension_pattern)[0][2].toLowerCase() : "bam"
+    def extension_matcher =  (args2 =~ extension_pattern)
+    def extension = extension_matcher.getCount() > 0 ? extension_matcher[0][2].toLowerCase() : "bam"
+    def reference = fasta && extension=="cram"  ? "--reference ${fasta}" : ""
+    if (!fasta && extension=="cram") error "Fasta reference is required for CRAM output"
 
     """
     INDEX=`find -L ./ -name "*.rev.1.bt2" | sed "s/\\.rev.1.bt2\$//"`
@@ -53,7 +61,7 @@ process BOWTIE2_ALIGN {
         $unaligned \\
         $args \\
         2> >(tee ${prefix}.bowtie2.log >&2) \\
-        | samtools $samtools_command $args2 --threads $task.cpus -o ${prefix}.${extension} -
+        | samtools $samtools_command $args2 --threads $task.cpus ${reference} -o ${prefix}.${extension} -
 
     if [ -f ${prefix}.unmapped.fastq.1.gz ]; then
         mv ${prefix}.unmapped.fastq.1.gz ${prefix}.unmapped_1.fastq.gz
@@ -82,9 +90,19 @@ process BOWTIE2_ALIGN {
     } else {
         create_unmapped = save_unaligned ? "touch ${prefix}.unmapped_1.fastq.gz && touch ${prefix}.unmapped_2.fastq.gz" : ""
     }
+    def reference = fasta && extension=="cram"  ? "--reference ${fasta}" : ""
+    if (!fasta && extension=="cram") error "Fasta reference is required for CRAM output"
+
+    def create_index = ""
+    if (extension == "cram") {
+        create_index = "touch ${prefix}.crai"
+    } else if (extension == "bam") {
+        create_index = "touch ${prefix}.csi"
+    }
 
     """
     touch ${prefix}.${extension}
+    ${create_index}
     touch ${prefix}.bowtie2.log
     ${create_unmapped}
 

diff --git a/modules/nf-core/bowtie2/align/tests/cram_crai.config b/modules/nf-core/bowtie2/align/tests/cram_crai.config
@@ -0,0 +1,5 @@
+process {
+    withName: BOWTIE2_ALIGN {
+        ext.args2 = '--output-fmt cram --write-index'
+    }
+}