Create a "genome file" that bedtools jaccard needs.

Otherwise it crashes if a sequence from B is not in A.

There is obvious repetition of the same computation here.  It is not
satisfying but I need a short term fix.  Hopefully it will put
the genome sequence in the node's memory cache, which will be useful
in the final commands.

modules/local/merge_repeatmasker_all.nf

@@ -24,8 +24,9 @@ process MERGE_REPM_RESULTS {
  def args = task.ext.args ?: ''
  def prefix = task.ext.prefix ?: "${meta.id}"
  """
+ awk '/^>/ {if (seqlen){print seqname "\t" seqlen}; split(\$1, a, ">"); seqname=a[2]; seqlen=0; next} {seqlen += length(\$0)} END {print seqname "\t" seqlen}' $genome > genome.genome # thanks, ChatGPT!
  run_bedtools_operations() {
- bedtools jaccard -nonamecheck -a "\$1" -b "\$2"   > "${prefix}_\${3}_jaccard.txt"
+ bedtools jaccard -nonamecheck -a "\$1" -b "\$2" -g genome.genome > "${prefix}_\${3}_jaccard.txt"
  zcat "\$1" "\$2" | sort -k1,1 -k2,2n | bedtools merge | gzip --best --no-name > "${prefix}_\${3}.mask.bed.gz"
  }
 

modules/local/merge_repeatmasker_dfam.nf

@@ -24,8 +24,9 @@ process MERGE_REPM_RESULTS {
  def args = task.ext.args ?: ''
  def prefix = task.ext.prefix ?: "${meta.id}"
  """
+ awk '/^>/ {if (seqlen){print seqname "\t" seqlen}; split(\$1, a, ">"); seqname=a[2]; seqlen=0; next} {seqlen += length(\$0)} END {print seqname "\t" seqlen}' $genome > genome.genome # thanks, ChatGPT!
  run_bedtools_operations() {
- bedtools jaccard -nonamecheck -a "\$1" -b "\$2"   > "${prefix}_\${3}_jaccard.txt"
+ bedtools jaccard -nonamecheck -a "\$1" -b "\$2" -g genome.genome > "${prefix}_\${3}_jaccard.txt"
  zcat "\$1" "\$2" | sort -k1,1 -k2,2n | bedtools merge | gzip --best --no-name > "${prefix}_\${3}.mask.bed.gz"
  }
 

modules/local/merge_repeatmasker_extlib.nf

@@ -24,8 +24,9 @@ process MERGE_REPM_RESULTS {
  def args = task.ext.args ?: ''
  def prefix = task.ext.prefix ?: "${meta.id}"
  """
+ awk '/^>/ {if (seqlen){print seqname "\t" seqlen}; split(\$1, a, ">"); seqname=a[2]; seqlen=0; next} {seqlen += length(\$0)} END {print seqname "\t" seqlen}' $genome > genome.genome # thanks, ChatGPT!
  run_bedtools_operations() {
- bedtools jaccard -nonamecheck -a "\$1" -b "\$2"   > "${prefix}_\${3}_jaccard.txt"
+ bedtools jaccard -nonamecheck -a "\$1" -b "\$2" -g genome.genome > "${prefix}_\${3}_jaccard.txt"
  zcat "\$1" "\$2" | sort -k1,1 -k2,2n | bedtools merge | gzip --best --no-name > "${prefix}_\${3}.mask.bed.gz"
  }
 

modules/local/mergemasks.nf

@@ -23,9 +23,10 @@ process MERGE_MASKS {
  def args = task.ext.args ?: ''
  def prefix = task.ext.prefix ?: "${meta.id}"
  """
- bedtools jaccard -nonamecheck -a $tantan -b $windowmasker > ${prefix}_tantan_windowmasker_jaccard.txt
- bedtools jaccard -nonamecheck -a $tantan -b $repeatmasker > ${prefix}_tantan_repeatmasker_jaccard.txt
- bedtools jaccard -nonamecheck -a $repeatmasker -b $windowmasker > ${prefix}_repeatmasker_windowmasker_jaccard.txt
+ awk '/^>/ {if (seqlen){print seqname "\t" seqlen}; split(\$1, a, ">"); seqname=a[2]; seqlen=0; next} {seqlen += length(\$0)} END {print seqname "\t" seqlen}' $genome > genome.genome # thanks, ChatGPT!
+ bedtools jaccard -nonamecheck -a $tantan -b $windowmasker -g genome.genome > ${prefix}_tantan_windowmasker_jaccard.txt
+ bedtools jaccard -nonamecheck -a $tantan -b $repeatmasker -g genome.genome > ${prefix}_tantan_repeatmasker_jaccard.txt
+ bedtools jaccard -nonamecheck -a $repeatmasker -b $windowmasker -g genome.genome > ${prefix}_repeatmasker_windowmasker_jaccard.txt
 
  zcat $tantan $windowmasker | sort -k1,1 -k2,2n | bedtools merge | gzip --best --no-name > ${prefix}_tantan_windowmasker.bed.gz
  zcat $tantan $repeatmasker | sort -k1,1 -k2,2n | bedtools merge | gzip --best --no-name > ${prefix}_tantan_repeatmasker.bed.gz