comparison defuse.xml @ 19:1af6f32ff592

Add datamanager, move to defuse_reference.loc

18:547d8db4673e

  <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"/>
  <conditional name="refGenomeSource">
      <param name="genomeSource" type="select" label="Will you select a built-in DeFuse Reference Dataset, or supply a configuration from your history" help="">
        <option value="indexed">Use a built-in DeFuse Reference Dataset</option>
        <option value="history">Use a configuration from your history that specifies the DeFuse Reference Dataset</option>
      </param>
      <when value="indexed">
        <param name="index" type="select" label="Select a Reference Dataset" help="if your genome of interest is not listed - contact Galaxy team">
          <options from_file="defuse.loc">
            <column name="name" index="1"/>
            <column name="value" index="2"/>
            <filter type="sort_by" column="0" />
            <validator type="no_options" message="No indexes are available" />
          </options>
        </param>
        <conditional name="defuse_param">
          <param name="settings" type="select" label="Defuse parameter settings" help="">
            <option value="preSet">Default settings</option>
            <option value="full">Full parameter list</option>
          </param>
          <when value="preSet" />
          <when value="full">
            <param name="max_insert_size" type="integer" value="500" optional="true" label="Bowtie max_insert_size" />
            <param name="dna_concordant_length" type="integer" value="2000" optional="true" label="Minimum gene fusion range dna_concordant_length" />
            <param name="discord_read_trim" type="integer" value="50" optional="true" label="Trim length for discordant reads discord_read_trim" help="(split reads are not trimmed)" />
            <param name="clustering_precision" type="float" value=".95" optional="true" label="Filter clustering_precision">
              <validator type="in_range" message="Choose a value between .1 and 1.0" min=".1" max="1"/>
            </param>
            <param name="span_count_threshold" type="integer" value="5" optional="true" label="Filter span_count_threshold" />
            <param name="split_count_threshold" type="integer" value="3" optional="true" label="Filter split_count_threshold" />
            <param name="percent_identity_threshold" type="float" value=".90" optional="true" label="Filter percent_identity_threshold">
              <validator type="in_range" message="Choose a value between .1 and 1.0" min=".1" max="1"/>
            </param>
            <param name="max_dist_pos" type="integer" value="600" optional="true" label="Filter max_dist_pos" />
            <param name="num_dist_genes" type="integer" value="500" optional="true" label="Filter num_dist_genes" />
            <param name="split_min_anchor" type="integer" value="4" optional="true" label="Filter split_min_anchor" />
            <param name="max_concordant_ratio" type="float" value="0.1" optional="true" label="Filter max_concordant_ratio">
              <validator type="in_range" message="Choose a value between 0.0 and 1.0" min="0" max="1"/>
            </param>
            <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>
            <!--
              <param name="positive_controls" type="data" format="txt" optional=true label="Defuse positive_controls" help=""/>
            -->
          </when> <!-- full -->
        </conditional>  <!-- defuse_param -->
      </when>
      <when value="history">
        <param name="config" type="data" format="txt" label="Defuse Config file" help=""/>
      </when>  <!-- history -->
  </conditional>  <!-- refGenomeSource -->
  <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="true" truevalue="yes" falsevalue="no" label="Run get_reads on each cluster"/>
 </inputs>

  <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 ${on_string}: defuse_output (purge when no longer needed)">
    <filter>keep_output == True</filter>
  </data>
  <data format="tabular" name="results_tsv" label="${tool.name} on ${on_string}: results.tsv" />
  <data format="tabular" name="results_classify_tsv" label="${tool.name} on ${on_string}: results.classify.tsv" />
  <data format="tabular" name="results_filtered_tsv" label="${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>
 </outputs>
 <configfiles>
  <configfile name="defuse_config">
#import ast
#if $refGenomeSource.genomeSource == "history":
#include raw $refGenomeSource.config.__str__
#else 
#set $ref_dict = dict($ast.literal_eval($refGenomeSource.index.value))
#
# Configuration file for defuse
#
# At a minimum, change all values enclused by []
#

# Directory where the defuse code was unpacked
## Default location in the tool/defuse directory  
# source_directory = ${__root_dir__}/tools/defuse
source_directory = #slurp
#try
$ref_dict['source_directory']
#except
__DEFUSE_PATH__
#end try

# Directory where you want your dataset
dataset_directory = #slurp
#try
$ref_dict['dataset_directory']

#except
\$(dataset_directory)/Hs.seq.uniq
#end try

# Paths to external tools
bowtie_bin = #slurp
#try
$ref_dict['bowtie_bin']
#except
__BOWTIE_BIN__
#end try
bowtie_build_bin = #slurp
#try
$ref_dict['bowtie_build_bin']
#except
__BOWTIE_BUILD_BIN__
#end try
blat_bin = #slurp
#try
$ref_dict['blat_bin']
#except
__BLAT_BIN__
#end try
fatotwobit_bin = #slurp
#try
$ref_dict['fatotwobit_bin']
#except
__FATOTWOBIT_BIN__
#end try
gmap_bin = #slurp
#try
$ref_dict['gmap_bin']
#except
__GMAP_BIN__
#end try
gmap_bin = #slurp
#try
$ref_dict['gmap_bin']
#except
__GMAP_BIN__
#end try
gmap_setup_bin = #slurp
#try
$ref_dict['gmap_setup_bin']
#except
__GMAP_SETUP_BIN__
#end try
r_bin = #slurp
#try
$ref_dict['r_bin']
#except
__R_BIN__
#end try
rscript_bin = #slurp
#try
$ref_dict['rscript_bin']
#except
__RSCRIPT_BIN__
#end try

# Directory where you want your dataset
gmap_index_directory = #slurp
#try
$ref_dict['gmap_index_directory']

$ref_dict['bowtie_quals']
#except
--phred33-quals
#end try
max_insert_size = #slurp
#if $refGenomeSource.defuse_param.settings == "full" and $refGenomeSource.defuse_param.max_insert_size.__str__ != "":
$refGenomeSource.defuse_param.max_insert_size
#else
#try
$ref_dict['max_insert_size']
#except
500

10000
#end try

# Minimum gene fusion range
dna_concordant_length = #slurp
#if $refGenomeSource.defuse_param.settings == "full" and $refGenomeSource.defuse_param.dna_concordant_length.__str__ != "":
$refGenomeSource.defuse_param.dna_concordant_length
#else
#try
$ref_dict['dna_concordant_length']
#except
2000
#end try
#end if

# Trim length for discordant reads (split reads are not trimmed)
discord_read_trim = #slurp
#if $refGenomeSource.defuse_param.settings == "full" and $refGenomeSource.defuse_param.discord_read_trim.__str__ != "":
$refGenomeSource.defuse_param.discord_read_trim
#else
#try
$ref_dict['discord_read_trim']
#except
50
#end try
#end if

# Filtering parameters
clustering_precision = #slurp
#if $refGenomeSource.defuse_param.settings == "full" and $refGenomeSource.defuse_param.clustering_precision.__str__ != ""
$refGenomeSource.defuse_param.clustering_precision
#else
#try
$ref_dict['clustering_precision']
#except
0.95
#end try
#end if
span_count_threshold = #slurp
#if $refGenomeSource.defuse_param.settings == "full" and $refGenomeSource.defuse_param.span_count_threshold.__str__ != ""
$refGenomeSource.defuse_param.span_count_threshold
#else
#try
$ref_dict['span_count_threshold']
#except
5
#end try
#end if
split_count_threshold = #slurp
#if $refGenomeSource.defuse_param.settings == "full" and $refGenomeSource.defuse_param.split_count_threshold.__str__ != ""
$refGenomeSource.defuse_param.split_count_threshold
#else
#try
$ref_dict['split_count_threshold']
#except
3
#end try
#end if
percent_identity_threshold = #slurp
#if $refGenomeSource.defuse_param.settings == "full" and $refGenomeSource.defuse_param.percent_identity_threshold.__str__ != ""
$refGenomeSource.defuse_param.percent_identity_threshold
#else
#try
$ref_dict['percent_identity_threshold']
#except
0.90
#end try
#end if
max_dist_pos = #slurp
#if $refGenomeSource.defuse_param.settings == "full" and $refGenomeSource.defuse_param.max_dist_pos.__str__ != ""
$refGenomeSource.defuse_param.max_dist_pos
#else
#try
$ref_dict['max_dist_pos']
#except
600
#end try
#end if
num_dist_genes = #slurp
#if $refGenomeSource.defuse_param.settings == "full" and $refGenomeSource.defuse_param.num_dist_genes.__str__ != ""
$refGenomeSource.defuse_param.num_dist_genes
#else
#try
$ref_dict['num_dist_genes']
#except
500
#end try
#end if
split_min_anchor = #slurp
#if $refGenomeSource.defuse_param.settings == "full" and $refGenomeSource.defuse_param.split_min_anchor.__str__ != ""
$refGenomeSource.defuse_param.split_min_anchor
#else
#try
$ref_dict['split_min_anchor']
#except
4
#end try
#end if
max_concordant_ratio = #slurp
#if $refGenomeSource.defuse_param.settings == "full" and $refGenomeSource.defuse_param.max_concordant_ratio.__str__ != ""
$refGenomeSource.defuse_param.max_concordant_ratio
#else
#try
$ref_dict['max_concordant_ratio']
#except
0.1
#end try
#end if
splice_bias = #slurp
#if $refGenomeSource.defuse_param.settings == "full" and $refGenomeSource.defuse_param.splice_bias.__str__ != ""
$refGenomeSource.defuse_param.splice_bias
#else
#try
$ref_dict['splice_bias']
#except
10
#end try
#end if
denovo_assembly = #slurp
#if $refGenomeSource.defuse_param.settings == "full" and $refGenomeSource.defuse_param.denovo_assembly.__str__ != ""
$refGenomeSource.defuse_param.denovo_assembly
#else
#try
$ref_dict['denovo_assembly']
#except
no
#end try
#end if
probability_threshold = #slurp
#if $refGenomeSource.defuse_param.settings == "full" and $refGenomeSource.defuse_param.probability_threshold.__str__ != ""
$refGenomeSource.defuse_param.probability_threshold
#else
#try
$ref_dict['probability_threshold']
#except
0.50

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

# Position density when calculating covariance
covariance_sampling_density = #slurp
#if $refGenomeSource.defuse_param.settings == "full" and $refGenomeSource.defuse_param.covariance_sampling_density.__str__ != ""
$refGenomeSource.defuse_param.covariance_sampling_density
#else
#try
$ref_dict['covariance_sampling_density']
#except
0.01
#end try
#end if


# Number of reads for each job in split
reads_per_job                               = 1000000

# Number of regions for each breakpoint sequence job in split
regions_per_job                             = 20

#raw
# If you have command line 'mail' and wish to be notified
# mailto                                      = [e-mail]

# Remove temp files
remove_job_files                            = yes
remove_job_temp_files                       = yes

# Converting to fastq
# Fastq converter config format 1 for reads stored in separate files for each end
#  data_lane_rexex_N is a perl regex which stores the lane id in $1
#  data_end_regex_N is a perl regex which stores the end, 1 or 2, in $1
#  data_compress_regex_N is a perl regex which stores the compression extension in $1
#  data_convert_N is the associated conversion utility that takes data at stdin and outputs fastq at stdout
# Fastq converter config format 2 for reads stored in separate files for each end
#  data_lane_regex_N is a perl regex which stores the lane id in $1
#  data_compress_regex_N is a perl regex which stores the compression extension in $1
#  data_end1_converter_N is the associated conversion utility that takes data at stdin and outputs fastq for end 1 at stdout
#  data_end2_converter_N is the associated conversion utility that takes data at stdin and outputs fastq for end 2 at stdout

data_lane_regex_1                           = ^(.+)_[12]_export\.txt.*$
data_end_regex_1                            = ^.+_([12])_export\.txt.*$
data_compress_regex_1                       = ^.+_[12]_export\.txt(.*)$
data_converter_1                            = $(scripts_directory)/fq_all2std.pl export2std

data_lane_regex_2                           = ^(.+)_[12]_concat_qseq\.txt.*$
data_end_regex_2                            = ^.+_([12])_concat_qseq\.txt.*$
data_compress_regex_2                       = ^.+_[12]_concat_qseq\.txt(.*)$
data_converter_2                            = $(scripts_directory)/qseq2fastq.pl

data_lane_regex_3                           = ^(.+)\.bam.*$
data_compress_regex_3                       = ^.+\.bam(.*)$
data_end1_converter_3                       = samtools view - | filter_sam_mate.pl 1 | sam_to_fastq.pl
data_end2_converter_3                       = samtools view - | filter_sam_mate.pl 2 | sam_to_fastq.pl

data_lane_regex_4                           = ^(.+).[12].fastq.*$
data_end_regex_4                            = ^.+.([12]).fastq.*$
data_compress_regex_4                       = ^.+.[12].fastq(.*)$
data_converter_4                            = cat
#end raw

#end if

  </configfile>
  <configfile name="shscript">
#!/bin/bash
## define some things for cheetah proccessing

#end if
## run defuse.pl
perl \${DEFUSE_PATH}/scripts/defuse.pl -c $defuse_config -1 data_dir/reads_1.fastq -2 data_dir/reads_2.fastq -o output_dir  -p 8
## 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
## create html with links for output_dir
#if $defuse_out.__str__ != 'None':
if [ -e $defuse_out ]

DeFuse requires 2 fastq files for paried reads, one with the left mate of the paired reads, and a second fastq with the the right mate of the paired reads (**with reads in the same order as in the first fastq dataset**).   

If your fastq files have reads in different orders or include unpaired reads,  you can preprocess them with **FASTQ interlacer** to create a single interlaced fastq dataset with only the paired reads and input that to **FASTQ de-interlacer** to separate the reads into a left fastq and right fastq.

DeFuse uses a Reference Dataset to search for gene fusions.  The Reference Dataset is generated from the following sources in DeFuse_Version_0.6_:
    - genome_fasta from Ensembl 
    - gene_models from Ensembl 
    - repeats_filename from UCSC RepeatMasker rmsk.txt
    - est_fasta from UCSC
    - est_alignments from UCSC intronEst.txt
    - unigene_fasta from NCBI

.. _DeFuse_Version_0.6: http://sourceforge.net/apps/mediawiki/defuse/index.php?title=DeFuse_Version_0.6.1

------

**Outputs**

19:1af6f32ff592

  <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"/>
  <conditional name="refGenomeSource">
    <param name="genomeSource" type="select" label="Will you select a built-in DeFuse Reference Dataset, or supply a configuration from your history" help="">
      <option value="indexed">Use a built-in DeFuse Reference Dataset</option>
      <option value="history">Use a configuration from your history that specifies the DeFuse Reference Dataset</option>
    </param>
    <when value="indexed">
      <param name="index" type="select" label="Select a Reference Dataset" help="if your genome of interest is not listed - contact Galaxy team">
        <options from_file="defuse_reference.loc">
          <column name="name" index="1"/>
          <column name="value" index="2"/>
          <filter type="sort_by" column="0" />
          <validator type="no_options" message="No indexes are available" />
        </options>
      </param>
    </when>
    <when value="history">
      <param name="config" type="data" format="defuse.conf" label="Defuse Config file" help=""/>
    </when>  <!-- history -->
  </conditional>  <!-- refGenomeSource -->
  <conditional name="defuse_param">
    <param name="settings" type="select" label="Defuse parameter settings" help="">
      <option value="preSet">Default settings</option>
      <option value="full">Full parameter list</option>
    </param>
    <when value="preSet" />
    <when value="full">
      <param name="max_insert_size" type="integer" value="500" optional="true" label="Bowtie max_insert_size" />
      <param name="dna_concordant_length" type="integer" value="2000" optional="true" label="Minimum gene fusion range dna_concordant_length" />
      <param name="discord_read_trim" type="integer" value="50" optional="true" label="Trim length for discordant reads discord_read_trim" help="(split reads are not trimmed)" />
      <param name="calculate_extra_annotations" type="select" label="Calculate extra annotations, fusion splice index and interrupted index" help="">
        <option value="">Use Default</option>
        <option value="no">no</option>
        <option value="yes">yes</option>
      </param>
      <param name="clustering_precision" type="float" value=".95" optional="true" label="Filter clustering_precision">
        <validator type="in_range" message="Choose a value between .1 and 1.0" min=".1" max="1"/>
      </param>
      <param name="span_count_threshold" type="integer" value="5" optional="true" label="Filter span_count_threshold" />
      <param name="percent_identity_threshold" type="float" value=".90" optional="true" label="Filter percent_identity_threshold">
        <validator type="in_range" message="Choose a value between .1 and 1.0" min=".1" max="1"/>
      </param>
      <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>
      <!--
        <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="do_get_reads" type="boolean" checked="true" truevalue="yes" falsevalue="no" label="Run get_reads on each cluster"/>
 </inputs>

  <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 ${on_string}: defuse_output (purge when no longer needed)">
    <filter>keep_output == True</filter>
  </data>
  <data format="tabular" name="results_classify_tsv" label="${tool.name} on ${on_string}: results.classify.tsv" />
  <data format="tabular" name="results_filtered_tsv" label="${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>
  <!--
   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__
#set 
#else 
#set config_file = $refGenomeSource.index.value
#end if
#set pat = '^\s*([^#=][^=]*?)\s*=\s*(.*?)\s*$'
#set fh = open()
#set keys = ['dataset_directory','ensembl_organism','ensembl_prefix','ensembl_version','ensembl_genome_version','ucsc_genome_version','ncbi_organism','ncbi_prefix','chromosomes','mt_chromosome','gene_sources','ig_gene_sources','rrna_gene_sources']
#set kv = []
#for $line in $fh:
  #set m = $re.match($pat,$line)
  #if $m and len($m.groups()) == 2:
    ## #echo $line
    #if $m.groups()[0] in keys:
      #set k = $m.groups()[0]
      #if k == 'dataset_directory' and $refGenomeSource.genomeSource == "indexed":
        ## The DataManager is conifgured to place the config file in the same directory as the defuse_data: dataset_directory
        #set v = $os.path.dirname($config_file)
      #else:
        #set v = $m.groups()[1]
      #end if
      #set kv = $kv + [[$k, $v]]
    #end if
  #end if
#end for
## #echo $kv
#set ref_dict = dict($kv)
## #echo $ref_dict
## include raw $refGenomeSource.config.__str__
#
# Configuration file for defuse
#
# At a minimum, change all values enclused by []
#

# Directory where the defuse code was unpacked
## Default location in the tool/defuse directory  
# source_directory = ${__root_dir__}/tools/defuse
source_directory = __DEFUSE_PATH__

# Directory where you want your dataset
dataset_directory = #slurp
#try
$ref_dict['dataset_directory']

#except
\$(dataset_directory)/Hs.seq.uniq
#end try

# Paths to external tools
bowtie_bin = __BOWTIE_BIN__
bowtie_build_bin = __BOWTIE_BUILD_BIN__
blat_bin = __BLAT_BIN__
fatotwobit_bin = __FATOTWOBIT_BIN__
gmap_bin = __GMAP_BIN__
gmap_bin = __GMAP_BIN__
gmap_setup_bin = __GMAP_SETUP_BIN__
r_bin = __R_BIN__
rscript_bin = __RSCRIPT_BIN__

# Directory where you want your dataset
gmap_index_directory = #slurp
#try
$ref_dict['gmap_index_directory']

$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
$ref_dict['max_insert_size']
#except
500

10000
#end try

# Minimum gene fusion range
dna_concordant_length = #slurp
#if $defuse_param.settings == "full" and $defuse_param.dna_concordant_length.__str__ != "":
$defuse_param.dna_concordant_length
#else
#try
$ref_dict['dna_concordant_length']
#except
2000
#end try
#end if

# Trim length for discordant reads (split reads are not trimmed)
discord_read_trim = #slurp
#if $defuse_param.settings == "full" and $defuse_param.discord_read_trim.__str__ != "":
$defuse_param.discord_read_trim
#else
#try
$ref_dict['discord_read_trim']
#except
50
#end try
#end if
# Calculate extra annotations, fusion splice index and interrupted index
calculate_extra_annotations = #slurp
#if $defuse_param.settings == "full" and $defuse_param.calculate_extra_annotations.__str__ != "":
$defuse_param.calculate_extra_annotations
#else
#try
$ref_dict['calculate_extra_annotations']
#except
no
#end try
#end if
# Filtering parameters
clustering_precision = #slurp
#if $defuse_param.settings == "full" and $defuse_param.clustering_precision.__str__ != ""
$defuse_param.clustering_precision
#else
#try
$ref_dict['clustering_precision']
#except
0.95
#end try
#end if
span_count_threshold = #slurp
#if $defuse_param.settings == "full" and $defuse_param.span_count_threshold.__str__ != ""
$defuse_param.span_count_threshold
#else
#try
$ref_dict['span_count_threshold']
#except
5
#end try
#end if
percent_identity_threshold = #slurp
#if $defuse_param.settings == "full" and $defuse_param.percent_identity_threshold.__str__ != ""
$defuse_param.percent_identity_threshold
#else
#try
$ref_dict['percent_identity_threshold']
#except
0.90
#end try
#end if
split_min_anchor = #slurp
#if $defuse_param.settings == "full" and $defuse_param.split_min_anchor.__str__ != ""
$defuse_param.split_min_anchor
#else
#try
$ref_dict['split_min_anchor']
#except
4
#end try
#end if
splice_bias = #slurp
#if $defuse_param.settings == "full" and $defuse_param.splice_bias.__str__ != ""
$defuse_param.splice_bias
#else
#try
$ref_dict['splice_bias']
#except
10
#end try
#end if
denovo_assembly = #slurp
#if $defuse_param.settings == "full" and $defuse_param.denovo_assembly.__str__ != ""
$defuse_param.denovo_assembly
#else
#try
$ref_dict['denovo_assembly']
#except
no
#end try
#end if
probability_threshold = #slurp
#if $defuse_param.settings == "full" and $defuse_param.probability_threshold.__str__ != ""
$defuse_param.probability_threshold
#else
#try
$ref_dict['probability_threshold']
#except
0.50

#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
#try
$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
#try
$ref_dict['reads_per_job']
#except
1000000
#end try
#end if

#raw
# If you have command line 'mail' and wish to be notified
# mailto                                      = [e-mail]

# 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

#end if
## run defuse.pl
perl \${DEFUSE_PATH}/scripts/defuse.pl -c $defuse_config -1 data_dir/reads_1.fastq -2 data_dir/reads_2.fastq -o output_dir  -p 8
## 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
## create html with links for output_dir
#if $defuse_out.__str__ != 'None':
if [ -e $defuse_out ]

DeFuse requires 2 fastq files for paried reads, one with the left mate of the paired reads, and a second fastq with the the right mate of the paired reads (**with reads in the same order as in the first fastq dataset**).   

If your fastq files have reads in different orders or include unpaired reads,  you can preprocess them with **FASTQ interlacer** to create a single interlaced fastq dataset with only the paired reads and input that to **FASTQ de-interlacer** to separate the reads into a left fastq and right fastq.

DeFuse uses a Reference Dataset to search for gene fusions.  The Reference Dataset is generated from the following sources in DeFuse_Version_0.4_:
    - genome_fasta from Ensembl 
    - gene_models from Ensembl 
    - repeats_filename from UCSC RepeatMasker rmsk.txt
    - est_fasta from UCSC
    - est_alignments from UCSC intronEst.txt
    - unigene_fasta from NCBI

.. _DeFuse_Version_0.4: http://sourceforge.net/apps/mediawiki/defuse/index.php?title=DeFuse_Version_0.4.2

------

**Outputs**