Fix FASTQ barcode parsing for non-standard header formats

fastq_to_ubam.nf

@@ -5,23 +5,41 @@ input_glob = params.input_glob ?: ['*.{1,2}.fastq.gz']
 read_format = params.read_format ?: 'paired-end'
 params.outdir = './ubam'
 
+// Shared shell function to extract and validate barcode from FASTQ header
+// Samples first 10k reads and returns the most frequent valid barcode
+def extractBarcodeFunction = '''
+extract_barcode() {
+    local fastq_file="$1"
+
+    # Extract last colon-field from comment (after space) of first 10k read headers
+    # Filter to valid barcodes (nucleotides with optional +), count occurrences, return most frequent
+    barcode=$(zcat "$fastq_file" \
+        | head -n 40000 \
+        | awk 'NR % 4 == 1 {sub(/.*[[:space:]]/, ""); n=split($0,a,":"); print a[n]}' \
+        | grep -E '^[ACGTN+-]+$' \
+        | sort | uniq -c | sort -rn | head -1 | awk '{print $2}')
+
+    # Fallback to unknown if no valid barcode found
+    echo "${barcode:-unknown}"
+}
+'''
+
 process FastqToBamPaired {
     conda "bioconda::picard=3.3.0 bioconda::samtools=1.21"
     publishDir "${params.outdir}", mode: 'copy'
     memory { params.max_memory ?: 300.GB }
-    
+
     input:
         tuple val(library), path(read1), path(read2)
-    
+
     output:
         path('*.bam')
 
     script:
     """
     set +o pipefail
-
-    barcode=\$(zcat ${read1} | head -n 1 | cut -d ":" -f 10)
-
+    ${extractBarcodeFunction}
+    barcode=\$(extract_barcode "${read1}")
     set -o pipefail
 
     picard FastqToSam TMP_DIR=/state/partition1/sge_tmp F1=${read1} F2=${read2} OUTPUT=temp.bam SM=${library} LB=${library} CN="New England Biolabs" PU=Illumina QUIET=true
@@ -35,19 +53,18 @@ process FastqToBamSingle {
     conda "bioconda::picard=3.3.0 bioconda::samtools=1.21"
     publishDir "${params.outdir}", mode: 'copy'
     memory { params.max_memory ?: 300.GB }
-    
+
     input:
         tuple val(library), path(read1)
-    
+
     output:
         path('*.bam')
 
     script:
     """
     set +o pipefail
-
-    barcode=\$(zcat ${read1} | head -n 1 | cut -d ":" -f 10)
-
+    ${extractBarcodeFunction}
+    barcode=\$(extract_barcode "${read1}")
     set -o pipefail
 
     picard FastqToSam F1=${read1} OUTPUT=temp.bam SM=${library} LB=${library} CN="New England Biolabs" PU=Illumina QUIET=true

tests/fastq_to_ubam.nf.test

@@ -40,4 +40,29 @@ nextflow_pipeline {
         )
         }
     }
+
+    test("fastq to uBam workflow - non-standard SRA headers (issue #15)") {
+        // Test that FASTQ files with SRA-style headers (e.g., @SRR... length=111)
+        // are handled gracefully without breaking due to spaces in extracted barcode
+        when {
+            params.input_glob = "$projectDir/tests/fixtures/fastq_sra/sra-test{.1,.2}.fastq.gz"
+        }
+
+        then {
+            def ubam = bam("${launchDir}/ubam/sra-test.bam").getStatistics()
+            // Read the BAM header to verify barcode is set to 'unknown' (not containing spaces)
+            def bamHeader = bam("${launchDir}/ubam/sra-test.bam").getHeader()
+            def rgLine = bamHeader.find { it.startsWith('@RG') }
+
+            assertAll(
+                { assert workflow.success },
+                // Verify the barcode field doesn't contain spaces (which would break commands)
+                { assert rgLine.contains('BC:unknown') : "Expected BC:unknown for non-standard headers, got: ${rgLine}" },
+                { assert snapshot(workflow.trace,
+                    ["ubam", ubam],
+                ).match()
+            }
+        )
+        }
+    }
 }

tests/fastq_to_ubam.nf.test.snap

@@ -25,6 +25,32 @@
         },
         "timestamp": "2025-08-20T09:23:01.410361919"
     },
+    "fastq to uBam workflow - non-standard SRA headers (issue #15)": {
+        "content": [
+            {
+                "tasksFailed": 0,
+                "tasksCount": 1,
+                "tasksSucceeded": 1
+            },
+            [
+                "ubam",
+                {
+                    "maxReadLength": 150,
+                    "minReadLength": 150,
+                    "meanReadLength": 150,
+                    "maxQuality": 0,
+                    "minQuality": 0,
+                    "meanQuality": 0,
+                    "readCount": 200
+                }
+            ]
+        ],
+        "meta": {
+            "nf-test": "0.9.3",
+            "nextflow": "25.10.2"
+        },
+        "timestamp": "2026-01-25T16:55:47.816983"
+    },
     "fastq to uBam workflow - paired-end": {
         "content": [
             {