diff variant_recalibrator.xml @ 4:f244b8209eb8 draft

bug fix release

--- a/variant_recalibrator.xml	Wed Feb 19 04:39:38 2014 -0500
+++ b/variant_recalibrator.xml	Mon Aug 25 17:43:11 2014 -0400
@@ -1,6 +1,8 @@
-<tool id="gatk2_variant_recalibrator" name="Variant Recalibrator" version="0.0.7">
+<tool id="gatk2_variant_recalibrator" name="Variant Recalibrator" version="@VERSION@.0">
   <description></description>
-  <expand macro="requirements" />
+  <expand macro="requirements">
+    <requirement type="package" version="0.9.3">ggplot</requirement>
+  </expand>
   <macros>
     <import>gatk2_macros.xml</import>
   </macros>
@@ -63,15 +65,12 @@
         --maxIterations "${analysis_param_type.max_iterations}"
         --numKMeans "${analysis_param_type.num_k_means}"
         --stdThreshold "${analysis_param_type.std_threshold}"
-        --qualThreshold "${analysis_param_type.qual_threshold}"
         --shrinkage "${analysis_param_type.shrinkage}"
         --dirichlet "${analysis_param_type.dirichlet}"
         --priorCounts "${analysis_param_type.prior_counts}"
-        #if str( $analysis_param_type.bad_variant_selector.bad_variant_selector_type ) == 'percent':
-            --percentBadVariants "${analysis_param_type.bad_variant_selector.percent_bad_variants}"
-        #else:
-            --minNumBadVariants "${analysis_param_type.bad_variant_selector.min_num_bad_variants}"
-        #end if
+
+        --minNumBadVariants "${analysis_param_type.min_num_bad_variants}"
+
         --target_titv "${analysis_param_type.target_titv}"
         #for $tranche in [ $tranche.strip() for $tranche in str( $analysis_param_type.ts_tranche ).split( ',' ) if $tranche.strip() ]
             --TStranche "${tranche}"
@@ -83,7 +82,6 @@
             #end if
             --ignore_filter "${ignore_filter_name}"
         #end for
-        --ts_filter_level "${analysis_param_type.ts_filter_level}"
         '
     #end if
     
@@ -100,7 +98,7 @@
           <param name="input_variants" type="data" format="vcf" label="Variant file to recalibrate" />
         </repeat>
         <param name="ref_file" type="select" label="Using reference genome" help="-R,--reference_sequence &amp;lt;reference_sequence&amp;gt;">
-          <options from_data_table="gatk_picard_indexes">
+          <options from_data_table="gatk2_picard_indexes">
           <!--  <filter type="data_meta" key="dbkey" ref="variants[0].input_variants" column="dbkey"/> -->
           </options>
           <validator type="no_options" message="A built-in reference genome is not available for the build associated with the selected input file"/>
@@ -114,7 +112,7 @@
       </when>
     </conditional>
     
-    <repeat name="rod_bind" title="Binding for reference-ordered data" help="-resource,--resource &amp;lt;resource&amp;gt;">
+    <repeat name="rod_bind" title="Binding for reference-ordered data" help="-resource,--resource &amp;lt;resource&amp;gt;" min="2">
         <conditional name="rod_bind_type">
           <param name="rod_bind_type_selector" type="select" label="Binding Type">
             <option value="dbsnp" selected="True">dbSNP</option>
@@ -324,26 +322,16 @@
     <expand macro="gatk_param_type_conditional" />
     
     <expand macro="analysis_type_conditional">
-        <param name="max_gaussians" type="integer" label="maximum number of Gaussians to try during variational Bayes Algorithm" value="10" help="-mG,--maxGaussians &amp;lt;maxGaussians&amp;gt;"/>
-        <param name="max_iterations" type="integer" label="maximum number of maximum number of VBEM iterations to be performed in variational Bayes Algorithm" value="100" help="-mI,--maxIterations &amp;lt;maxIterations&amp;gt;"/>
-        <param name="num_k_means" type="integer" label="number of k-means iterations to perform in order to initialize the means of the Gaussians in the Gaussian mixture model" value="30" help="-nKM,--numKMeans &amp;lt;numKMeans&amp;gt;"/>
-        <param name="std_threshold" type="float" label="If a variant has annotations more than -std standard deviations away from mean then don't use it for building the Gaussian mixture model." value="8.0" help="-std,--stdThreshold &amp;lt;stdThreshold&amp;gt;"/>
-        <param name="qual_threshold" type="float" label="If a known variant has raw QUAL value less than -qual then don't use it for building the Gaussian mixture model." value="80.0" help="-qual,--qualThreshold &amp;lt;qualThreshold&amp;gt;"/>
+        <param name="max_gaussians" type="integer" label="maximum number of Gaussians to try during variational Bayes Algorithm" value="8" help="-mG,--maxGaussians &amp;lt;maxGaussians&amp;gt;"/>
+        <param name="max_iterations" type="integer" label="maximum number of maximum number of VBEM iterations to be performed in variational Bayes Algorithm" value="150" help="-mI,--maxIterations &amp;lt;maxIterations&amp;gt;"/>
+        <param name="num_k_means" type="integer" label="number of k-means iterations to perform in order to initialize the means of the Gaussians in the Gaussian mixture model" value="100" help="-nKM,--numKMeans &amp;lt;numKMeans&amp;gt;"/>
+        <param name="std_threshold" type="float" label="If a variant has annotations more than -std standard deviations away from mean then don't use it for building the Gaussian mixture model." value="10.0" help="-std,--stdThreshold &amp;lt;stdThreshold&amp;gt;"/>
         <param name="shrinkage" type="float" label="shrinkage parameter in variational Bayes algorithm" value="1.0" help="-shrinkage,--shrinkage &amp;lt;shrinkage&amp;gt;"/>
         <param name="dirichlet" type="float" label="dirichlet parameter in variational Bayes algorithm" value="0.001" help="-dirichlet,--dirichlet &amp;lt;dirichlet&amp;gt;"/>
         <param name="prior_counts" type="float" label="number of prior counts to use in variational Bayes algorithm" value="20.0" help="-priorCounts,--priorCounts &amp;lt;priorCounts&amp;gt;"/>
-        <conditional name="bad_variant_selector">
-          <param name="bad_variant_selector_type" type="select" label="How to specify bad variants">
-            <option value="percent" selected="True">Percent</option>
-            <option value="min_num">Number</option>
-          </param>
-          <when value="percent">
-            <param name="percent_bad_variants" type="float" label="percentage of the worst scoring variants to use when building the Gaussian mixture model of bad variants. 0.07 means bottom 7 percent." value="0.03" help="-percentBad,--percentBadVariants &amp;lt;percentBadVariants&amp;gt;"/>
-          </when>
-          <when value="min_num">
-            <param name="min_num_bad_variants" type="integer" label="minimum amount of worst scoring variants to use when building the Gaussian mixture model of bad variants. Will override -percentBad arugment if necessary" value="2000" help="-minNumBad,--minNumBadVariants &amp;lt;minNumBadVariants&amp;gt;"/>
-          </when>
-        </conditional>
+        <!--<param name="trustAllPolymorphic" type="boolean" label="trustAllPolymorphic" truevalue="-/-trustAllPolymorphic=true" falsevalue="-/-trustAllPolymorphic=false"
+            help="Trust that all the input training sets' unfiltered records contain only polymorphic sites to drastically speed up the computation. -trustAllPolymorphic" />-->
+        <param name="min_num_bad_variants" type="integer" label="Minimum number of worst scoring variants to use when building the Gaussian mixture model of bad variants" value="1000" help="--minNumBadVariants &amp;lt;minNumBadVariants&amp;gt;"/>
         <param name="target_titv" type="float" label="expected novel Ti/Tv ratio to use when calculating FDR tranches and for display on optimization curve output figures. (approx 2.15 for whole genome experiments). ONLY USED FOR PLOTTING PURPOSES!" value="2.15" help="-titv,--target_titv &amp;lt;target_titv&amp;gt;"/>
         <param name="ts_tranche" type="text" label="levels of novel false discovery rate (FDR, implied by ti/tv) at which to slice the data. (in percent, that is 1.0 for 1 percent)" value="100.0, 99.9, 99.0, 90.0" help="-tranche,--TStranche &amp;lt;TStranche&amp;gt;"/>
         <repeat name="ignore_filters" title="Ignore Filter" help="-ignoreFilter,--ignore_filter &amp;lt;ignore_filter&amp;gt;">
@@ -360,7 +348,6 @@
             <when value="LowQual" />
           </conditional>
         </repeat>
-        <param name="ts_filter_level" type="float" label="truth sensitivity level at which to start filtering, used here to indicate filtered variants in plots" value="99.0" help="-ts_filter_level,--ts_filter_level &amp;lt;ts_filter_level&amp;gt;"/>
     </expand>
   </inputs>
   <outputs>
@@ -410,12 +397,10 @@
  maxIterations         The maximum number of VBEM iterations to be performed in variational Bayes algorithm. Procedure will normally end when convergence is detected.
  numKMeans             The number of k-means iterations to perform in order to initialize the means of the Gaussians in the Gaussian mixture model.
  stdThreshold          If a variant has annotations more than -std standard deviations away from mean then don't use it for building the Gaussian mixture model.
- qualThreshold         If a known variant has raw QUAL value less than -qual then don't use it for building the Gaussian mixture model.
  shrinkage             The shrinkage parameter in variational Bayes algorithm.
  dirichlet             The dirichlet parameter in variational Bayes algorithm.
  priorCounts           The number of prior counts to use in variational Bayes algorithm.
- percentBadVariants    What percentage of the worst scoring variants to use when building the Gaussian mixture model of bad variants. 0.07 means bottom 7 percent.
- minNumBadVariants     The minimum amount of worst scoring variants to use when building the Gaussian mixture model of bad variants. Will override -percentBad arugment if necessary.
+ minNumBadVariants     The minimum amount of worst scoring variants to use when building the Gaussian mixture model of bad variants.
  recal_file            The output recal file used by ApplyRecalibration
  target_titv           The expected novel Ti/Tv ratio to use when calculating FDR tranches and for display on optimization curve output figures. (approx 2.15 for whole genome experiments). ONLY USED FOR PLOTTING PURPOSES!
  TStranche             The levels of novel false discovery rate (FDR, implied by ti/tv) at which to slice the data. (in percent, that is 1.0 for 1 percent)
@@ -423,8 +408,8 @@
  path_to_Rscript       The path to your implementation of Rscript. For Broad users this is maybe /broad/tools/apps/R-2.6.0/bin/Rscript
  rscript_file          The output rscript file generated by the VQSR to aid in visualization of the input data and learned model
  path_to_resources     Path to resources folder holding the Sting R scripts.
- ts_filter_level       The truth sensitivity level at which to start filtering, used here to indicate filtered variants in plots
 
 @CITATION_SECTION@
   </help>
+  <expand macro="citations" />
 </tool>