Commit b2915bcc authored by Christoph Ziegenhain's avatar Christoph Ziegenhain
Browse files

Merge branch 'zUMIs-dev' of https://github.com/sdparekh/zUMIs into zUMIs-dev

parents 21ba6c57 97b1789a

listPrefix.sh

+7 −0
Original line number Diff line number Diff line
@@ -8,6 +8,10 @@ zumisdir=$1 
tmpMerge=$2

f=$3

project=$4

yaml=$5

samtoolsexc=$6

Rexc=$7

pigzexc=$8



if [[ $f =~ \.gz$ ]]; then

	pref=`basename $f .gz`
@@ -16,3 +20,6 @@ else 
	pref=`basename $f`

	ls $tmpMerge$pref* | sed "s|$tmpMerge$pref||" > $tmpMerge/$project.listPrefix.txt

fi



for x in `cat $tmpMerge/$project.listPrefix.txt`; do perl $zumisdir/fqfilter_v2.pl $yaml $samtoolsexc $Rexc $pigzexc $zumisdir $x & done

wait

splitfq.sh

+12 −4
Original line number Diff line number Diff line
@@ -45,8 +45,16 @@ function splitfqgz() { 
i=$1

pigzexc=$2

num_threads=$3

nlines=$4

tmpMerge=$5

fun=$6

project=$7

tmpMerge=$4

fun=$5

project=$6

f=$7





if [[ $f =~ \.gz$ ]]; then

	nlines=`pigz -p $num_threads -d -c $f | wc -l`

else

	nlines=`wc -l $f | awk '{print $1}'`

fi



$fun "$i" "$pigzexc" "$num_threads" "$nlines" "$tmpMerge" "$project"

zUMIs-master.sh

+12 −19
Original line number Diff line number Diff line
@@ -135,10 +135,6 @@ if 
[[ "$whichStage" == "Filtering" ]]

then

  echo "Starting Filtering..."

  #echo "samtools_exec: $samtoolsexc" >> $yaml

  #echo "pigz_exec: $pigzexc" >> $yaml

  #echo "STAR_exec: $starexc" >> $yaml

  #echo "zUMIs_directory: $zumisdir" >> $yaml



  f=`cut -d' ' -f1 <(echo $fqfiles)` # the first fastq file to determine gzip status
@@ -147,37 +143,34 @@ then 
  if [[ "$isslurm" == "yes" ]]; then



    if [[ $f =~ \.gz$ ]]; then

      nlines=`pigz -p $num_threads -d -c $f | wc -l`

      for i in $fqfiles;do sbatch --cpus-per-task=1 --mem=10M --wrap="bash $zumisdir/splitfq.sh $i $pigzexc $num_threads $nlines $tmpMerge splitfqgz $project" > $outdir/$project.splitfq.slurmjobid.txt;done

      echo "splitfq"

      for i in $fqfiles;do sbatch --cpus-per-task=1 --job-name=splitfq --mem=10M --wrap="bash $zumisdir/splitfq.sh $i $pigzexc $num_threads $tmpMerge splitfqgz $project $f" > $outdir/$project.splitfq.slurmjobid.txt;done

    else

      nlines=`wc -l $f | awk '{print $1}'`

      for i in $fqfiles;do sbatch --cpus-per-task=1 --mem=10M --wrap="bash $zumisdir/splitfq.sh $i $pigzexc $num_threads $nlines $tmpMerge splitfq $project" > $outdir/$project.splitfq.slurmjobid.txt;done

      for i in $fqfiles;do sbatch --cpus-per-task=1 --job-name=splitfq --mem=10M --wrap="bash $zumisdir/splitfq.sh $i $pigzexc $num_threads $tmpMerge splitfq $project $f" > $outdir/$project.splitfq.slurmjobid.txt;done

    fi



    j=`cat $outdir/$project.splitfq.slurmjobid.txt | cut -f4 -d' '`



    sbatch --cpus-per-task=1 --dependency=afterok:'$j' --mem=1M --wrap="bash $zumisdir/listPrefix.sh $zumisdir $tmpMerge $f $project" > $outdir/$project.listPrefix.slurmjobid.txt

    sbatch --cpus-per-task=1 --job-name=listPrefix --dependency=afterok:'$j' --mem=1M --wrap="bash $zumisdir/listPrefix.sh $zumisdir $tmpMerge $f $project $yaml $samtoolsexc $Rexc $pigzexc" > $outdir/$project.listPrefix.slurmjobid.txt



    j=`cat $outdir/$project.listPrefix.slurmjobid.txt | cut -f4 -d' '`



    for x in `cat $tmpMerge/$project.listPrefix.txt`;do sbatch --cpus-per-task=$num_threads --dependency=afterok:'$j' --mem=10M --wrap="bash $zumisdir/fqfilter_v2.pl $yaml $samtoolsexc $Rexc $pigzexc $zumisdir $x" > $outdir/$project.fqfilter_v2.slurmjobid.txt;done

  #  for x in `cat $tmpMerge/$project.listPrefix.txt`;do sbatch --cpus-per-task=$num_threads --dependency=afterok:'$j' --mem=10M --wrap="bash $zumisdir/fqfilter_v2.pl $yaml $samtoolsexc $Rexc $pigzexc $zumisdir $x" > $outdir/$project.fqfilter_v2.slurmjobid.txt;done



    j=`cat $outdir/$project.fqfilter_v2.slurmjobid.txt | cut -f4 -d' '`

  #  j=`cat $outdir/$project.fqfilter_v2.slurmjobid.txt | cut -f4 -d' '`



    sbatch --cpus-per-task=1 --dependency=afterok:'$j' --mem=1M --wrap="bash $zumisdir/mergeBAM.sh $zumisdir $tmpMerge $num_threads $project $outdir $yaml" > $outdir/$project.mergeBAM.slurmjobid.txt

    sbatch --cpus-per-task=1 --job-name=splitfq=mergeBAM --dependency=afterok:'$j' --mem=1M --wrap="bash $zumisdir/mergeBAM.sh $zumisdir $tmpMerge $num_threads $project $outdir $yaml" > $outdir/$project.mergeBAM.slurmjobid.txt



    j=`cat $outdir/$project.mergeBAM.slurmjobid.txt | cut -f4 -d' '`



  else



    if [[ $f =~ \.gz$ ]]; then

      nlines=`pigz -p $num_threads -d -c $f | wc -l`

      for i in $fqfiles;do bash $zumisdir/splitfq.sh $i $pigzexc $num_threads $nlines $tmpMerge splitfqgz $project;done

      for i in $fqfiles;do bash $zumisdir/splitfq.sh $i $pigzexc $num_threads $tmpMerge splitfqgz $project $f;done

      pref=`basename $f .gz`

      l=`ls $tmpMerge$pref* | sed "s|$tmpMerge$pref||" | sed 's/.gz//'`

    else

      nlines=`wc -l $f | awk '{print $1}'`

      for i in $fqfiles;do bash $zumisdir/splitfq.sh $i $pigzexc $num_threads $nlines $tmpMerge splitfq $project;done

      for i in $fqfiles;do bash $zumisdir/splitfq.sh $i $pigzexc $num_threads $tmpMerge splitfq $project $f;done

      pref=`basename $f`

      l=`ls $tmpMerge$pref* | sed "s|$tmpMerge$pref||"`

    fi
@@ -196,7 +189,7 @@ then 
  if [[ "$isslurm" == "yes" ]]; then

    memory=`du -sh $genomedir | cut -f1` #STAR genome index size

    j=`cat $outdir/$project.mergeBAM.slurmjobid.txt | cut -f4 -d' '`

    sbatch --dependency=afterok:'$j' --mem=$memory --cpus-per-task=$num_threads --wrap="$Rexc $zumisdir/zUMIs-mapping.R $yaml" > $outdir/$project.mapping.slurmjobid.txt

    sbatch --dependency=afterok:'$j' --job-name=mapping --mem=$memory --cpus-per-task=$num_threads --wrap="$Rexc $zumisdir/zUMIs-mapping.R $yaml" > $outdir/$project.mapping.slurmjobid.txt

  else

    $Rexc $zumisdir/zUMIs-mapping.R $yaml

  fi
@@ -216,7 +209,7 @@ then 
      mem_limit=`expr $mem_limit \* 1000`

    fi

    j=`cat $outdir/$project.mapping.slurmjobid.txt | cut -f4 -d' '`

    sbatch --dependency=afterok:'$j' --mem=$mem_limit --cpus-per-task=$num_threads --wrap="$Rexc $zumisdir/zUMIs-dge2.R $yamlnew" > $outdir/$project.dge.slurmjobid.txt

    sbatch --dependency=afterok:'$j' --mem=$mem_limit --job-name=dge --cpus-per-task=$num_threads --wrap="$Rexc $zumisdir/zUMIs-dge2.R $yamlnew" > $outdir/$project.dge.slurmjobid.txt

   else

     $Rexc $zumisdir/zUMIs-dge2.R $yamlnew

   fi
@@ -233,7 +226,7 @@ then 
    echo "Starting descriptive statistics..."

    if [[ "$isslurm" == "yes" ]]; then

      j=`cat $outdir/$project.dge.slurmjobid.txt | cut -f4 -d' '`

      sbatch --dependency=afterok:'$j' --mem=5000 --cpus-per-task=$num_threads --wrap="$Rexc $zumisdir/zUMIs-stats2.R $yamlnew" > $outdir/$project.stats.slurmjobid.txt

      sbatch --dependency=afterok:'$j' --mem=5000 --job-name=stats --cpus-per-task=$num_threads --wrap="$Rexc $zumisdir/zUMIs-stats2.R $yamlnew" > $outdir/$project.stats.slurmjobid.txt

    else

      $Rexc $zumisdir/zUMIs-stats2.R $yamlnew

    fi

CRA Git | Maintained and supported by SUSTech CRA and CCSE