Add alignment input possibility for rsem and tidy up tests

modules/nf-core/rsem/calculateexpression/main.nf

@@ -8,14 +8,14 @@ process RSEM_CALCULATEEXPRESSION {
         'community.wave.seqera.io/library/rsem_star:5acb4e8c03239c32' }"
 
     input:
-    tuple val(meta), path(reads)
+    tuple val(meta), path(reads)  // FASTQ files or BAM file for --alignments mode
     path  index
 
     output:
     tuple val(meta), path("*.genes.results")   , emit: counts_gene
     tuple val(meta), path("*.isoforms.results"), emit: counts_transcript
     tuple val(meta), path("*.stat")            , emit: stat
-    tuple val(meta), path("*.log")             , emit: logs
+    tuple val(meta), path("*.log")             , emit: logs, optional:true
     path  "versions.yml"                       , emit: versions
 
     tuple val(meta), path("*.STAR.genome.bam")       , optional:true, emit: bam_star
@@ -35,14 +35,34 @@ process RSEM_CALCULATEEXPRESSION {
     } else if (meta.strandedness == 'reverse') {
         strandedness = '--strandedness reverse'
     }
-    def paired_end = meta.single_end ? "" : "--paired-end"
+
+    // Detect if input is BAM file(s)
+    def is_bam = reads.toString().toLowerCase().endsWith('.bam')
+    def alignment_mode = is_bam ? '--alignments' : ''
+
+    // Use metadata for paired-end detection if available, otherwise empty (auto-detect)
+    def paired_end = meta.containsKey('single_end') ? (meta.single_end ? "" : "--paired-end") : "unknown"
+
     """
     INDEX=`find -L ./ -name "*.grp" | sed 's/\\.grp\$//'`
+
+    # Use metadata-based paired-end detection, or auto-detect if no metadata provided
+    PAIRED_END_FLAG="$paired_end"
+    if [ "${paired_end}" == "unknown" ]; then
+        # Auto-detect only if no metadata provided
+        if [ "${is_bam}" == "true" ]; then
+            samtools flagstat $reads | grep -q 'paired in sequencing' && PAIRED_END_FLAG="--paired-end"
+        else
+            [ ${reads.size()} -gt 1 ] && PAIRED_END_FLAG="--paired-end"
+        fi
+    fi
+
     rsem-calculate-expression \\
         --num-threads $task.cpus \\
         --temporary-folder ./tmp/ \\
+        $alignment_mode \\
         $strandedness \\
-        $paired_end \\
+        \$PAIRED_END_FLAG \\
         $args \\
         $reads \\
         \$INDEX \\
@@ -57,12 +77,18 @@ process RSEM_CALCULATEEXPRESSION {
 
     stub:
     prefix = task.ext.prefix ?: "${meta.id}"
+    def is_bam = reads.toString().toLowerCase().endsWith('.bam')
     """
     touch ${prefix}.genes.results
     touch ${prefix}.isoforms.results
     touch ${prefix}.stat
     touch ${prefix}.log
-    touch ${prefix}.STAR.genome.bam
+
+    # Only create STAR BAM output when not in alignment mode
+    if [ "${is_bam}" == "false" ]; then
+        touch ${prefix}.STAR.genome.bam
+    fi
+
     touch ${prefix}.genome.bam
     touch ${prefix}.transcript.bam
 

modules/nf-core/rsem/calculateexpression/meta.yml

@@ -21,10 +21,11 @@ input:
           e.g. [ id:'test', single_end:false ]
     - reads:
         type: file
-        description: Input reads for quantification
-        pattern: "*.fastq.gz"
+        description: Input reads for quantification (FASTQ files or BAM file for --alignments mode)
+        pattern: "*.{fastq.gz,bam}"
         ontologies:
           - edam: http://edamontology.org/format_3989 # GZIP format
+          - edam: http://edamontology.org/format_2572 # BAM format
   - index:
       type: file
       description: RSEM index

modules/nf-core/rsem/calculateexpression/tests/alignment.config

@@ -0,0 +1,5 @@
+process {
+
+    publishDir = { "${params.outdir}/${task.process.tokenize(':')[-1].tokenize('_')[0].toLowerCase()}" }
+
+}

modules/nf-core/rsem/calculateexpression/tests/main.nf.test

@@ -3,14 +3,59 @@ nextflow_process {
     name "Test Process RSEM_CALCULATEEXPRESSION"
     script "../main.nf"
     process "RSEM_CALCULATEEXPRESSION"
-    config "./nextflow.config"
     tag "modules"
     tag "modules_nfcore"
     tag "rsem"
     tag "rsem/calculateexpression"
     tag "rsem/preparereference"
 
-    test("homo_sapiens") {
+
+    test("homo_sapiens - bam") {
+
+        config "./alignment.config"
+
+        setup {
+            run("RSEM_PREPAREREFERENCE") {
+                script "../../../rsem/preparereference/main.nf"
+                process {
+                    """
+                    input[0] = Channel.of(file(params.modules_testdata_base_path + 'genomics/homo_sapiens/genome/genome.fasta', checkIfExists: true))
+                    input[1] = Channel.of(file(params.modules_testdata_base_path + 'genomics/homo_sapiens/genome/genome.gtf', checkIfExists: true))
+                    """
+                }
+            }
+        }
+
+        when {
+            params {
+                outdir = "$outputDir"
+            }
+            process {
+                """
+                input[0] = Channel.of([
+                    [ id:'test', strandedness: 'forward' ], // meta map
+                    file(params.modules_testdata_base_path + 'genomics/homo_sapiens/illumina/bam/rsem.transcript.bam', checkIfExists: true)
+                ])
+                input[1] = RSEM_PREPAREREFERENCE.out.index
+                """
+            }
+        }
+        then {
+            assertAll(
+                { assert process.success },
+                { assert snapshot(
+                    process.out.counts_gene,
+                    process.out.counts_transcript,
+                    process.out.stat,
+                    process.out.versions
+                ).match() }
+            )
+        }
+    }
+
+    test("homo_sapiens - fastq") {
+
+        config "./nextflow.config"
 
         setup {
             run("RSEM_PREPAREREFERENCE") {
@@ -44,18 +89,22 @@ nextflow_process {
         then {
             assertAll(
                 { assert process.success },
-                { assert snapshot(process.out.counts_gene).match("counts_gene") },
-                { assert snapshot(process.out.counts_transcript).match("counts_transcript") },
-                { assert snapshot(process.out.stat).match("stat") },
-                { assert path(process.out.logs.get(0).get(1)).exists() },
-                { assert snapshot(process.out.versions).match("versions") }
+                { assert snapshot(
+                    process.out.counts_gene,
+                    process.out.counts_transcript,
+                    process.out.stat,
+                    process.out.versions
+                ).match() },
+                { assert path(process.out.logs.get(0).get(1)).exists() }
             )
         }
     }
 
     test("homo_sapiens - stub") {
 
-    options "-stub"
+        config "./nextflow.config"
+
+        options "-stub"
 
         setup {
             run("RSEM_PREPAREREFERENCE") {
@@ -86,7 +135,13 @@ nextflow_process {
         then {
             assertAll(
                 { assert process.success },
-                { assert snapshot(process.out).match() }
+                { assert path(process.out.counts_gene.get(0).get(1)).exists() },
+                { assert path(process.out.counts_transcript.get(0).get(1)).exists() },
+                { assert path(process.out.stat.get(0).get(1)).exists() },
+                { assert path(process.out.logs.get(0).get(1)).exists() },
+                { assert snapshot(
+                    process.out.versions
+                ).match() }
             )
         }
     }