comparison defuse.xml @ 45:aedaa66483f1 draft

Uploaded

44:225750bf3770

    <macros>
        <import>macros.xml</import>
    </macros>
    <requirements>
        <expand macro="defuse_requirement" />
        <expand macro="mapping_requirements" />
        <expand macro="r_requirements" />
    </requirements>
  <command interpreter="command"> /bin/bash $shscript </command>
 <inputs>
  <param name="left_pairendreads" type="data" format="fastq" label="left part of read pairs" help="The left and right reads pairs must be in the same order, and not have any unpaired reads.  (FASTQ interlacer will pair reads and remove the unpaired.   FASTQ de-interlacer will separate the result into left and right reads.)"/>
  <param name="right_pairendreads" type="data" format="fastq" label="right part of read pairs" help="In the same order as the left reads"/>
  <param name="library_name" type="text" value="unknown" label="library name" help="Value to put in the results library_name column">
    <validator type="length" min="1"/>

      <param name="split_min_anchor" type="integer" value="4" optional="true" label="Filter split_min_anchor" />
      <param name="splice_bias" type="integer" value="10" optional="true" label="Filter splice_bias" />
      <param name="probability_threshold" type="float" value="0.50" optional="true" label="Filter probability_threshold">
        <validator type="in_range" message="Choose a value between 0.0 and 1.0" min="0" max="1"/>
      </param>
      <param name="covariance_sampling_density" type="float" value="0.01" optional="true" label="covariance_sampling_density">
        <help>Position density when calculating covariance</help>
        <validator type="in_range" message="Choose a value between 0.0 and 1.0" min="0" max="1"/>
      </param>
      <param name="denovo_assembly" type="select" label="denovo_assembly" help="">
        <option value="">Use Default</option>
        <option value="no">no</option>
        <option value="yes">yes</option>

        <param name="positive_controls" type="data" format="txt" optional=true label="Defuse positive_controls" help=""/>
      -->
      <param name="reads_per_job" type="integer" value="1000000" optional="true" label="Number of reads for each job in split" />
    </when> <!-- full -->
  </conditional>  <!-- defuse_param -->
  <param name="breakpoints_bam" type="boolean" checked="true" truevalue="yes" falsevalue="no" label="Generate a Bam file for the fusions"/>
  <param name="keep_output" type="boolean" checked="true" truevalue="yes" falsevalue="no" label="Save DeFuse working directory files" 
         help="The defuse output working directory can be helpful for determining errors that may have occurred during the run, 
               but they require considerable diskspace, and should be deleted and purged when no longer needed."/>
  <param name="do_get_reads" type="boolean" checked="false" truevalue="yes" falsevalue="no" label="Run get_reads on each cluster"/>
 </inputs>
 <stdio>
   <exit_code range="1:"  level="fatal" description="Error Running Defuse" />
 </stdio>
 <outputs>
  <data format="txt" name="config_txt" label="${tool.name} on ${on_string}: config.txt"/>
  <data format="txt" name="defuse_log" label="${tool.name} on ${on_string}: defuse.log" />
  <data format="html" name="defuse_out" label="${tool.name} on ${on_string}: defuse_output (purge when no longer needed)">
    <filter>keep_output == True</filter>
  </data>
  <data format="defuse.results.tsv" name="results_classify_tsv" label="${tool.name} on ${on_string}: results.classify.tsv" />
  <data format="defuse.results.tsv" name="results_filtered_tsv" label="${tool.name} on ${on_string}: results.filtered.tsv" />
  <data format="html" name="fusion_reads" label="${tool.name} on ${on_string}: fusion_reads">
    <filter>do_get_reads == True</filter>
  </data>
  <data format="bam" name="fusions_bam" label="${tool.name} on ${on_string}: fusions.bam">
    <filter>breakpoints_bam == True</filter>
  </data>
  <!--
   expression_plot
   circos plot
  -->
 </outputs>
 <configfiles>
  <configfile name="defuse_config">
#import re
#set $ds = chr(36)
#if $refGenomeSource.genomeSource == "history":
#set config_file = $refGenomeSource.config.__str__
#else 
#set config_file = $refGenomeSource.index.value
#end if

#try
$ref_dict['bowtie_quals']
#except
--phred33-quals
#end try
max_insert_size = #slurp
#if $defuse_param.settings == "full" and $defuse_param.max_insert_size.__str__ != "":
$defuse_param.max_insert_size
#else
#try

0.50
#end try
#end if
positive_controls                           = \$(data_directory)/controls.txt

# Position density when calculating covariance
covariance_sampling_density = #slurp
#if $defuse_param.settings == "full" and $defuse_param.covariance_sampling_density.__str__ != ""
$defuse_param.covariance_sampling_density
#else

$ref_dict['covariance_sampling_density']
#except
0.01
#end try
#end if
# Number of reads for each job in split
reads_per_job = #slurp
#if $defuse_param.settings == "full" and $defuse_param.reads_per_job.__str__ != ""
$defuse_param.reads_per_job
#else

# Remove temp files
remove_job_files                            = yes
remove_job_temp_files                       = yes

#end raw


  </configfile>
  <configfile name="shscript">
#!/bin/bash
## define some things for cheetah proccessing
#set $ds = chr(36)
#set $amp = chr(38)
#set $gt = chr(62)
#set $lt = chr(60)
#set $echo_cmd = 'echo'
## Find the defuse.pl in the galaxy tool path
#import Cheetah.FileUtils
## declare a bash function for converting a results tsv into html with links to the get_reads output files
results2html() {
  rlts=${ds}1
  rslt_name=`basename ${ds}rlts`
  html=${ds}2
  echo '${lt}html${gt}${lt}head${gt}${lt}title${gt}Defuse '${ds}rslt_name'${lt}/title${gt}${lt}/head${gt}${lt}body${gt}' ${gt}  ${ds}html
  echo '${lt}h2${gt}Defuse '${ds}rslt_name'${lt}/h2${gt}${lt}table${gt}' ${gt}${gt}  ${ds}html
  if [ -z "${ds}3" ]  
  then
    awk '${ds}1 ~ /cluster_id/{printf("${lt}tr${gt}");for (i = 1; i ${lt}= NF; i++) {printf("${lt}th${gt}%s${lt}/th${gt}", ${ds}i);}; printf("${lt}/tr${gt}\n");}\
         ${ds}1 ~ /[1-9][0-9]*/{printf("${lt}tr${gt}");for (i = 1; i ${lt}= NF; i++) {printf("${lt}td${gt}%s${lt}/td${gt}", ${ds}i);}; printf("${lt}/tr${gt}\n");}' ${ds}rlts ${gt}${gt} ${ds}html
    echo '${lt}/table${gt}' ${gt}${gt} ${ds}html
    echo '${lt}/body${gt}${lt}/html${gt}' ${gt}${gt}  ${ds}html
  else
    export _EFP=${ds}3
    mkdir -p ${ds}_EFP
    awk '${ds}1 ~ /cluster_id/{printf("${lt}tr${gt}");for (i = 1; i ${lt}= NF; i++) {printf("${lt}th${gt}%s${lt}/th${gt}", ${ds}i);}; printf("${lt}/tr${gt}\n");}\
         ${ds}1 ~ /[1-9][0-9]*/{fn="cluster_"${ds}1"_reads.txt"; \
          printf("${lt}tr${gt}${lt}td${gt}${lt}a href=\"%s\"${gt}%s${lt}/a${gt}${lt}/td${gt}",fn, ${ds}1);for (i = 2; i ${lt}= NF; i++) {printf("${lt}td${gt}%s${lt}/td${gt}", ${ds}i);}; printf("${lt}/tr${gt}\n");}' ${ds}rlts ${gt}${gt} ${ds}html
    echo '${lt}/table${gt}' ${gt}${gt} ${ds}html
    echo '${lt}/body${gt}${lt}/html${gt}' ${gt}${gt}  ${ds}html
    for i in `awk '${ds}1 ~ /[1-9][0-9]*/{print ${ds}1}' ${ds}rlts`;
      do fn=cluster_${ds}{i}_reads.txt;
      pn=${ds}_EFP/${ds}fn;
      perl \${DEFUSE_PATH}/scripts/get_reads.pl -c $defuse_config -o output_dir -i ${ds}i ${gt} ${ds}pn;
    done
  fi
}
## substitute pathnames into config file
if `grep __DEFUSE_PATH__ $defuse_config ${gt} /dev/null`;then sed -i'.tmp' "s#__DEFUSE_PATH__#\${DEFUSE_PATH}#" $defuse_config; fi
if `grep __SAMTOOLS_BIN__ $defuse_config ${gt} /dev/null` ${amp}${amp} SAMTOOLS_BIN=`which samtools`;then sed -i'.tmp' "s#__SAMTOOLS_BIN__#\${SAMTOOLS_BIN}#" $defuse_config; fi
if `grep __BOWTIE_BIN__ $defuse_config ${gt} /dev/null` ${amp}${amp} BOWTIE_BIN=`which bowtie`;then sed -i'.tmp' "s#__BOWTIE_BIN__#\${BOWTIE_BIN}#" $defuse_config; fi
if `grep __BOWTIE_BUILD_BIN__ $defuse_config ${gt} /dev/null` ${amp}${amp} BOWTIE_BUILD_BIN=`which bowtie-build`;then sed -i'.tmp' "s#__BOWTIE_BUILD_BIN__#\${BOWTIE_BUILD_BIN}#" $defuse_config; fi
if `grep __BLAT_BIN__ $defuse_config ${gt} /dev/null` ${amp}${amp} BLAT_BIN=`which blat`;then sed -i'.tmp' "s#__BLAT_BIN__#\${BLAT_BIN}#" $defuse_config; fi
if `grep __FATOTWOBIT_BIN__ $defuse_config ${gt} /dev/null` ${amp}${amp} FATOTWOBIT_BIN=`which faToTwoBit`;then sed -i'.tmp' "s#__FATOTWOBIT_BIN__#\${FATOTWOBIT_BIN}#" $defuse_config; fi
if `grep __GMAP_BIN__ $defuse_config ${gt} /dev/null` ${amp}${amp} GMAP_BIN=`which gmap`;then sed -i'.tmp' "s#__GMAP_BIN__#\${GMAP_BIN}#" $defuse_config; fi
if `grep __GMAP_SETUP_BIN__ $defuse_config ${gt} /dev/null` ${amp}${amp} GMAP_SETUP_BIN=`which gmap_setup`;then sed -i'.tmp' "s#__GMAP_SETUP_BIN__#\${GMAP_SETUP_BIN}#" $defuse_config; fi
if `grep __R_BIN__ $defuse_config ${gt} /dev/null` ${amp}${amp} R_BIN=`which R`;then sed -i'.tmp' "s#__R_BIN__#\${R_BIN}#" $defuse_config; fi
if `grep __RSCRIPT_BIN__ $defuse_config ${gt} /dev/null` ${amp}${amp} RSCRIPT_BIN=`which Rscript`;then sed -i'.tmp' "s#__RSCRIPT_BIN__#\${RSCRIPT_BIN}#" $defuse_config; fi


## copy config to output
cp $defuse_config $config_txt
## make a data_dir  and ln -s the input fastq
mkdir -p data_dir
## ln -s "$left_pairendreads" data_dir/reads_1.fastq
## ln -s "$right_pairendreads" data_dir/reads_2.fastq
cp "$left_pairendreads" data_dir/reads_1.fastq
cp "$right_pairendreads" data_dir/reads_2.fastq
## ln to output_dir in from_work_dir
#if $defuse_out.__str__ != 'None':
mkdir -p $defuse_out.dataset.extra_files_path
ln -s $defuse_out.dataset.extra_files_path  output_dir
#else
mkdir -p output_dir
#end if
## run defuse.pl
perl \${DEFUSE_PATH}/scripts/defuse.pl -name "$library_name" -c $defuse_config -1 data_dir/reads_1.fastq -2 data_dir/reads_2.fastq -o output_dir  -p \$GALAXY_SLOTS
## copy primary results to output datasets
if [ -e output_dir/log/defuse.log ]; then cp output_dir/log/defuse.log $defuse_log; fi
## if [ -e output_dir/results.tsv ]; then cp output_dir/results.tsv $results_tsv; fi
if [ -e output_dir/results.filtered.tsv ]; then cp output_dir/results.filtered.tsv $results_filtered_tsv; fi
if [ -e output_dir/results.classify.tsv ]; then cp output_dir/results.classify.tsv $results_classify_tsv; fi
#if $breakpoints_bam:
if [ -e output_dir/results.filtered.tsv ] ${amp}${amp}  [ -e output_dir/breakpoints.genome.psl ]
then
  awk "\\$10 ~ /^(`awk '\\$1 ~ /[0-9]+/{print \\$1}' output_dir/results.filtered.tsv | tr '\n' '|'`)\\$/{print \\$0}" output_dir/breakpoints.genome.psl > breakpoints.genome.filtered.psl ${amp}${amp}
  psl2sam.pl breakpoints.genome.filtered.psl > breakpoints.genome.filtered.sam ${amp}${amp}
  samtools view -b -T /panfs/roc/rissdb/galaxy/genomes/NCBIM37/defuse/defuse.reference.fa -o breakpoints.genome.filtered.bam breakpoints.genome.filtered.sam ${amp}${amp}
  samtools sort breakpoints.genome.filtered.bam breakpoints ${amp}${amp}
  ## samtools index breakpoints.bam
  cp breakpoints.bam $fusions_bam
fi
#end if
## create html with links for output_dir
#if $defuse_out.__str__ != 'None':
if [ -e $defuse_out ]
then
  echo '${lt}html${gt}${lt}head${gt}${lt}title${gt}Defuse Output${lt}/title${gt}${lt}/head${gt}${lt}body${gt}' ${gt} $defuse_out
  echo '${lt}h2${gt}Defuse Output Files${lt}/h2${gt}${lt}ul${gt}' ${gt}${gt}  $defuse_out
  pushd $defuse_out.dataset.extra_files_path
  for f in `find -L . -maxdepth 1 -type f`; 
   do fn=`basename ${ds}f`; echo '${lt}li${gt}${lt}a href="'${ds}fn'"${gt}'${ds}fn'${lt}/a${gt}${lt}/li${gt}' ${gt}${gt}  $defuse_out; 
  done
  popd
  echo '${lt}/ul${gt}' ${gt}${gt} $defuse_out
  echo '${lt}/body${gt}${lt}/html${gt}' ${gt}${gt}  $defuse_out
fi
#end if
## run get_reads.pl on each cluster
#if $fusion_reads.__str__ != 'None':
if [ -e output_dir/results.filtered.tsv -a -e $fusion_reads ] 
then
  mkdir -p $fusion_reads.dataset.extra_files_path
  results2html output_dir/results.filtered.tsv $fusion_reads $fusion_reads.dataset.extra_files_path
fi
#end if
  </configfile>
 </configfiles>

 <tests>
 </tests>

45:aedaa66483f1

    <macros>
        <import>macros.xml</import>
    </macros>
    <requirements>
        <expand macro="defuse_requirement" />
    </requirements>
  <command><![CDATA[
    #if $defuse_out.__str__ != 'None':
        ## ln to output_dir in from_work_dir
        mkdir -p $defuse_out.dataset.extra_files_path &&
        ln -s $defuse_out.dataset.extra_files_path  output_dir &&
    #else
      mkdir -p output_dir &&
    #end if
    ## Put executable paths in config file
    $__tool_directory__/config_sub.sh $defuse_config output_dir/defuse.cfg &&
    ## copy config to output
    cp defuse.cfg $config_txt &&
    ## make a data_dir  and ln -s the input fastq
    mkdir -p data_dir &&
    ln -s "$left_pairendreads" data_dir/reads_1.fastq &&
    ln -s "$right_pairendreads" data_dir/reads_2.fastq &&
    ## run 
    perl defuse_run.pl --name "$library_name" --config defuse.cfg  -1 data_dir/reads_1.fastq -2 data_dir/reads_2.fastq -o output_dir  -p \$GALAXY_SLOTS &&
    grep -v cluster_id  output_dir/results.filtered.tsv | awk '{print $1}' > cluster_id_list && 
    get_fusion_fastq.pl --list cluster_id_list --output output_dir --fastq1 results.fusions_1.fq --fastq2 results.fusions_2.fq && 
    cp output_dir/results.* .  &&
    cp `find output_dir -name defuse.log` $defuse_log 
    #if $defuse_out.__str__ != 'None':
        && $__tool_directory__/make_html.sh $defuse_out $defuse_out.dataset.extra_files_path
    #end if
  ]]></command>
 <inputs>
  <param name="left_pairendreads" type="data" format="fastq" label="left part of read pairs" help="The left and right reads pairs must be in the same order, and not have any unpaired reads.  (FASTQ interlacer will pair reads and remove the unpaired.   FASTQ de-interlacer will separate the result into left and right reads.)"/>
  <param name="right_pairendreads" type="data" format="fastq" label="right part of read pairs" help="In the same order as the left reads"/>
  <param name="library_name" type="text" value="unknown" label="library name" help="Value to put in the results library_name column">
    <validator type="length" min="1"/>

      <param name="split_min_anchor" type="integer" value="4" optional="true" label="Filter split_min_anchor" />
      <param name="splice_bias" type="integer" value="10" optional="true" label="Filter splice_bias" />
      <param name="probability_threshold" type="float" value="0.50" optional="true" label="Filter probability_threshold">
        <validator type="in_range" message="Choose a value between 0.0 and 1.0" min="0" max="1"/>
      </param>
      <param name="multi_exon_transcripts_stats" type="select" label="Use multiple exon transcripts for stats calculations" help="should be enabled for very small libraries">
        <option value="no" select="true">no</option>
        <option value="yes">yes</option>
      </param>
      <param name="covariance_sampling_density" type="float" value="0.01" optional="true" label="covariance_sampling_density">
        <help>Position density when calculating covariance</help>
        <validator type="in_range" message="Choose a value between 0.0 and 1.0" min="0" max="1"/>
      </param>
      <param name="max_paired_alignments" type="integer" value="10" optional="true" label="max_paired_alignments">
        <help>Maximum number of alignments for a read pair, Pairs with more alignments are filtered, default is 10</help>
        <validator type="in_range" message="Choose a value between 0.0 and 1.0" min="1" max="100"/>
      </param>
      <param name="denovo_assembly" type="select" label="denovo_assembly" help="">
        <option value="">Use Default</option>
        <option value="no">no</option>
        <option value="yes">yes</option>

        <param name="positive_controls" type="data" format="txt" optional=true label="Defuse positive_controls" help=""/>
      -->
      <param name="reads_per_job" type="integer" value="1000000" optional="true" label="Number of reads for each job in split" />
    </when> <!-- full -->
  </conditional>  <!-- defuse_param -->
  <param name="keep_output" type="boolean" checked="true" truevalue="yes" falsevalue="no" label="Save DeFuse working directory files" 
         help="The defuse output working directory can be helpful for determining errors that may have occurred during the run, 
               but they require considerable diskspace, and should be deleted and purged when no longer needed."/>
  <param name="breakpoints_bam" type="boolean" checked="false" truevalue="yes" falsevalue="no" label="Generate a Bam file for the fusions"/>
  <param name="do_get_reads" type="boolean" checked="false" truevalue="yes" falsevalue="no" label="Run get_reads on each cluster"/>
 </inputs>
 <outputs>
  <data format="txt" name="config_txt" label="${tool.name} on ${on_string}: config.txt"/>
  <data format="txt" name="defuse_log" label="${tool.name} on ${on_string}: defuse.log" />
  <data format="html" name="defuse_out" label="${tool.name} on ${on_string}: defuse_output (purge when no longer needed)">
    <filter>keep_output == True</filter>
  </data>
  <data format="defuse.results.tsv" name="results_classify_tsv" label="${tool.name} on ${on_string}: results.classify.tsv" from_work_dir="results.classify.tsv"/>
  <data format="defuse.results.tsv" name="results_filtered_tsv" label="${tool.name} on ${on_string}: results.filtered.tsv" from_work_dir="results.filtered.tsv"/>
  <data format="fastqsanger" name="results_fusions1_fq" label="${tool.name} on ${on_string}: fusions_1.fq" from_work_dir="results.fusions_1.fq" />
  <data format="fastqsanger" name="results_fusions2_fq" label="${tool.name} on ${on_string}: fusions_2.fq" from_work_dir="results.fusions_2.fq" />
  <!--
   expression_plot
   circos plot
  -->
 </outputs>
 <configfiles>
  <configfile name="defuse_config">
#import re
#if $refGenomeSource.genomeSource == "history":
#set config_file = $refGenomeSource.config.__str__
#else 
#set config_file = $refGenomeSource.index.value
#end if

#try
$ref_dict['bowtie_quals']
#except
--phred33-quals
#end try
bowtie_params = #slurp
#try
$ref_dict['bowtie_params']
#except
--chunkmbs 200
#end try
max_insert_size = #slurp
#if $defuse_param.settings == "full" and $defuse_param.max_insert_size.__str__ != "":
$defuse_param.max_insert_size
#else
#try

0.50
#end try
#end if
positive_controls                           = \$(data_directory)/controls.txt

# Use multiple exon transcripts for stats calculations (yes/no)
# should be enabled for very small libraries
multi_exon_transcripts_stats = #slurp
#if $defuse_param.settings == "full" and $defuse_param.multi_exon_transcripts_stats.__str__ != ""
$defuse_param.multi_exon_transcripts_stats
#else
#try
$ref_dict['multi_exon_transcripts_stats']
#except
no
#end try
#end if

# Position density when calculating covariance
covariance_sampling_density = #slurp
#if $defuse_param.settings == "full" and $defuse_param.covariance_sampling_density.__str__ != ""
$defuse_param.covariance_sampling_density
#else

$ref_dict['covariance_sampling_density']
#except
0.01
#end try
#end if

# Maximum number of alignments for a read pair
# Pairs with more alignments are filtered
max_paired_alignments = #slurp
#if $defuse_param.settings == "full" and $defuse_param.max_paired_alignments.__str__ != ""
$defuse_param.max_paired_alignments
#else
#try
$ref_dict['max_paired_alignments']
#except
10
#end try
#end if

# Number of reads for each job in split
reads_per_job = #slurp
#if $defuse_param.settings == "full" and $defuse_param.reads_per_job.__str__ != ""
$defuse_param.reads_per_job
#else

# Remove temp files
remove_job_files                            = yes
remove_job_temp_files                       = yes

qsub_params                                 = ""

#end raw

  </configfile>
 </configfiles>

 <tests>
 </tests>