Hello,

I recently started using Nextflow and I always get this error:

WARN: Input tuple does not match input set cardinality declared by process markDuplicates -- offending value:

The code is as follows:

Channel
  .fromPath(params.samples)
  .splitCsv()
  .map { row ->
  def sampleID = row[0]
  def tag = row[1]
  def read1 = file(row[2])
  def read2 = file(row[3])

  return [ sampleID, tag, read1, read2 ]
  }
  .set { allSamples_ch }


  process align {

  publishDir "${params.out}/aligned_reads", mode:'copy'

  input:
  set val(sampleID), val(tag), file(read1), file(read2) from allSamples_ch

  output:
  set val(sampleID), val(tag), file("${sampleID}_${tag}.bam") into reads_ch

  script:
  readGroup = \
    "@RG\\tID:${sampleID}_${tag}\\tLB:${sampleID}_${tag}\\tPL:${params.pl}\\tSM:${sampleID}_${tag}"
  """
  $BWA mem -M -t 24 \
    -R \"${readGroup}\" \
    $ref \
    ${read1} \
    ${read2} | \
    $SAMTOOLS sort -@ 24 -o ${sampleID}_${tag}.bam -
  """
    }


  process markDuplicates {


  publishDir "${params.out}/dedup_sorted", mode:'copy'

  input:
  set val(sampleID), val(tag), file(aligned_reads) from reads_ch.collect()

  output:
  set val(sampleID), val(tag), file("${sampleID}_sorted_dedup_${tag}.bam") into bam_for_variant_calling, \
  sorted_dedup_ch_for_metrics, bam_for_indexing
  set val(sampleID), val(tag), file("${sampleID}_dedup_metrics_${tag}.txt") into dedup_qc_ch

  script:
  """
  $PICARD MarkDuplicates \
  I=${sampleID}_${tag}.bam \
  O=${sampleID}_sorted_dedup_${tag}.bam \
  M=${sampleID}_dedup_metrics_${tag}.txt \
  TMP_DIR=${params.tmpdir}/${workflow.runName}/${sampleID}
  """
  }

The align process was successful but I couldn't fix the markDuplicate step.

I will really appreciate your help folks!

Regards

Joe