main.nf

#!/usr/bin/env nextflow

params.strain = "Arab_me14"
params.reference = "data/${params.strain}/reference/*.fasta"
params.reads = "data/${params.strain}/illumina/pairs/*[12].fastq.gz"

reference = file(params.reference).first()

reads = Channel
.fromPath(params.reads)
.toSortedList()
.flatten()
.take(4)
.map { path -> 
  matcher = (path.getFileName() =~ /(.*)_R[12].fastq.gz/)
  matcher.matches()
  return [matcher.group(1), path]
}
.groupTuple(sort: true, size: 2)
.set { illuminaPairs }

process tophat2 {
  tag { runID }
  cache 'deep'
  cpus 4
  
  input:
  file reference
  set val(runID), file("reads.*.fastq.gz") from illuminaPairs

  output:
  set val(runID), file("tophat_out/accepted_hits.bam") into mappedPairs

  script:
  match = (runID =~ /(\d+_\d+).*/)
  match.matches()
  def sampleID = match.group(1)
  """
bowtie2-build $reference reference \
&& ln -s $reference reference.fa
tophat2 \
--num-threads ${task.cpus} \
--microexon-search \
--b2-very-fast \
--min-intron-length 5 \
--max-intron-length 200 \
--rg-id ${runID} \
--rg-sample '${sampleID}' \
--rg-library '${sampleID}' \
--rg-description 'Illumina sequencing run ${runID} for sample ${sampleID}' \
--rg-platform ILLUMINA \
--library-type fr-firststrand \
reference \
reads.1.fastq.gz \
reads.2.fastq.gz
  """
}

mappedPairs
.map { runID, bamfile ->
  matcher = (runID =~ /(\d+_\d+).*/)
  matcher.matches()
  return [matcher.group(1), bamfile]
 }
.groupTuple()
.set { unmergedBams1 }

process merge_to_runs {
  tag { sampleID }

  input:
  set val(sampleID), file("hits.*.bam") from unmergedBams1

  output:
  set val(sampleID), file("merged.bam") into sampleBams

  """
java -jar $baseDir/bin/picard.jar \
  MergeSamFiles \
  OUTPUT=merged.bam \
  ASSUME_SORTED=true \
  \$(printf ' INPUT=%s' hits.*.bam)
  """
}

sampleBams
.map { sampleID, bamfile ->
  matcher = (sampleID =~ /(\d+)_\d+/)
  matcher.matches()
  Integer sampleNum = matcher.group(1).toInteger()
  String timepoint = "unknown"
  switch (sampleNum) {
  case 11: timepoint = "demo";
  case [33, 41, 49]: timepoint = "early";
  case [34, 42, 50]: timepoint = "medium";
  case [35, 43, 51]: timepoint = "late";
  case [40, 48, 56]: timepoint = "culture";
  }

  return [timepoint, bamfile]
 }
.groupTuple()
.set { unmergedBams2 }

process merge_to_conditions {
  tag { timepoint }
  
  input:
  set val(timepoint), file("hits.*.bam") from unmergedBams2

  output:
  file("${params.strain}.${timepoint}.bam")

  """
java -jar $baseDir/bin/picard.jar \
  MergeSamFiles \
  OUTPUT=${params.strain}.${timepoint}.bam \
  ASSUME_SORTED=true \
  \$(printf ' INPUT=%s' hits.*.bam)
  """
}