Mercurial > repos > ebi-gxa > scanpy_find_variable_genes

--- a/scanpy-find-variable-genes.xml	Wed Mar 13 11:51:42 2019 -0400
+++ b/scanpy-find-variable-genes.xml	Thu Mar 12 09:12:26 2020 +0000
@@ -1,52 +1,56 @@
 <?xml version="1.0" encoding="utf-8"?>
-<tool id="scanpy_find_variable_genes" name="Scanpy FindVariableGenes" version="@TOOL_VERSION@+galaxy0">
+<tool id="scanpy_find_variable_genes" name="Scanpy FindVariableGenes" version="@TOOL_VERSION@+galaxy10" profile="@PROFILE@">
   <description>based on normalised dispersion of expression</description>
   <macros>
-    <import>scanpy_macros.xml</import>
+    <import>scanpy_macros2.xml</import>
   </macros>
   <expand macro="requirements"/>
   <command detect_errors="exit_code"><![CDATA[
 ln -s '${input_obj_file}' input.h5 &&
-PYTHONIOENCODING=utf-8 scanpy-find-variable-genes.py -i input.h5
-    -f '${input_format}'
-    -o output.h5
-    -F '${output_format}'
-    --flavor '${flavor}'
-    -b '${n_bin}'
-    #if $parameters
-        #set pars = ','.join([str($p['name']) for $p in $parameters])
-        -p '${pars}'
-        #set mins = ','.join([str($p['min']) for $p in $parameters])
-        -l '${mins}'
-        #set maxs = ','.join([str($p['max']) for $p in $parameters])
-        -j '${maxs}'
-    #end if
-    #if $n_top_gene
-        -n '${n_top_gene}'
-    #end if
+PYTHONIOENCODING=utf-8 scanpy-find-variable-genes
+    --flavor '${method.flavor}'
+#if $method.flavor == 'seurat'
+    --mean-limits ${method.min_mean} ${method.max_mean}
+    --disp-limits ${method.min_disp} ${method.max_disp}
+#else
+    --n-top-genes ${method.n_top_gene}
+#end if
+    --n-bins '${n_bin}'
+    ${filter}
+    @INPUT_OPTS@
+    @OUTPUT_OPTS@
 ]]></command>

   <inputs>
     <expand macro="input_object_params"/>
     <expand macro="output_object_params"/>
-    <param name="flavor" argument="--flavor" type="select" value="seurat" label="Flavor of computing normalised dispersion">
-      <option value="seurat">Seurat</option>
-      <option value="cell_ranger">Cell-ranger</option>
-    </param>
-    <repeat name="parameters" min="1" title="Parameters used to find variable genes">
-      <param name="name" type="select" label="Name of parameter to filter on">
-        <option value="mean">Mean of expression</option>
-        <option value="disp">Dispersion of expression</option>
+    <conditional name="method">
+      <param name="flavor" argument="--flavor" type="select" label="Flavor of computing normalised dispersion">
+        <option value="seurat" selected="true">Seurat</option>
+        <option value="cell_ranger">Cell-ranger</option>
       </param>
-      <param name="min" type="float" value="0" label="Min value"/>
-      <param name="max" type="float" value="1e9" label="Max value"/>
-    </repeat>
+      <when value="seurat">
+        <param name="min_mean" argument="--min-mean" type="float" min="0" value="0.0125"
+               label="Min value for normalised mean expression (in log1p scale)"/>
+        <param name="max_mean" argument="--max-mean" type="float" min="0" value="3"
+               label="Max value for normalised mean expresssion (in log1p scale)"/>
+        <param name="min_disp" argument="--min-disp" type="float" min="0" value="0.5"
+               label="Min value for dispersion of expression"/>
+        <param name="max_disp" argument="--max-disp" type="float" min="0" value="50"
+               label="Max value for dispersion of expresssion"/>
+      </when>
+      <when value="cell_ranger">
+        <param name="n_top_gene" argument="--n-top-genes" type="integer" value="2000"
+               label="Number of top variable genes to keep"/>
+      </when>
+    </conditional>
     <param name="n_bin" argument="--n-bins" type="integer" value="20" label="Number of bins for binning the mean expression"/>
-    <param name="n_top_gene" argument="--n-top-genes" type="integer" optional="true" label="Number of top variable genes to keep"/>
+    <param name="filter" argument="--subset" type="boolean" truevalue="--subset" falsevalue="" checked="false"
+           label="Remove genes not marked as highly variable"/>
   </inputs>

   <outputs>
-    <data name="output_h5" format="h5" from_work_dir="output.h5" label="${tool.name} on ${on_string}: Variable genes"/>
+    <expand macro="output_data_obj" description="Variable genes"/>
   </outputs>

   <tests>
@@ -56,26 +60,20 @@
       <param name="output_format" value="anndata"/>
       <param name="flavor" value="seurat"/>
       <param name="n_bin" value="20"/>
-      <repeat name="parameters">
-        <param name="name" value="mean"/>
-        <param name="min" value="0.0125"/>
-        <param name="max" value="3"/>
-      </repeat>
-      <repeat name="parameters">
-        <param name="name" value="disp"/>
-        <param name="min" value="0.5"/>
-        <param name="max" value="1e9"/>
-      </repeat>
+      <param name="min_mean" value="0.0125"/>
+      <param name="max_mean" value="3"/>
+      <param name="min_disp" value="0.5"/>
+      <param name="max_disp" value="1e9"/>
       <output name="output_h5" file="find_variable_genes.h5" ftype="h5" compare="sim_size"/>
     </test>
   </tests>

   <help><![CDATA[
-============================================================
-Extract highly variable genes (`pp.filter_genes_dispersion`)
-============================================================
+==============================================================
+Mark highly variable genes (`scanpy.pp.highly_variable_genes`)
+==============================================================

-Depending on `flavor`, this reproduces the R-implementations of Seurat and Cell Ranger.
+Depending on `flavor`, this reproduces the R-implementations of Seurat or Cell Ranger.

 The normalized dispersion is obtained by scaling with the mean and standard
 deviation of the dispersions for genes falling into a given bin for mean
--- a/scanpy_macros.xml	Wed Mar 13 11:51:42 2019 -0400
+++ b/scanpy_macros.xml	Thu Mar 12 09:12:26 2020 +0000
@@ -38,9 +38,14 @@
       <yield/>
     </requirements>
   </xml>
+  <token name="@EXPORT_MTX_OPTS@">
+      ${export_mtx}
+  </token>
   <token name="@VERSION_HISTORY@"><![CDATA[
 **Version history**

+1.3.2+galaxy1: Normalise-data and filter-genes: Exposes ability to output 10x files.
+
 1.3.2+galaxy0: Initial contribution. Ni Huang and Pablo Moreno, Expression Atlas team https://www.ebi.ac.uk/gxa/home  at
 EMBL-EBI https://www.ebi.ac.uk/ and Teichmann Lab at Wellcome Sanger Institute.
     ]]></token>
@@ -87,4 +92,18 @@
     <param name="sort_order" argument="--no-sort-order" type="boolean" checked="true" label="Element with high color-by value plot on top"/>
     <param name="frameoff" argument="--frameoff" type="boolean" checked="false" label="Omit frame"/>
   </xml>
+  <xml name="export_mtx_params">
+    <param name="export_mtx" argument="--export-mtx" type="boolean" truevalue="--export-mtx ./" falsevalue="" checked="false" label="Save normalised data to 10x format" help="If enabled, it will generate in addition to the main output in Loom or AnnData an export in 10x format of the normalised data."/>
+  </xml>
+  <xml name="export_mtx_outputs">
+    <data name="matrix_10x" format="txt" from_work_dir="matrix.mtx" label="${tool.name} on ${on_string}: 10x matrix">
+      <filter>export_mtx</filter>
+    </data>
+    <data name="genes_10x" format="tsv" from_work_dir="genes.tsv" label="${tool.name} on ${on_string}: 10x genes">
+      <filter>export_mtx</filter>
+    </data>
+    <data name="barcodes_10x" format="tsv" from_work_dir="barcodes.tsv" label="${tool.name} on ${on_string}: 10x barcodes">
+      <filter>export_mtx</filter>
+    </data>
+  </xml>
 </macros>
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/scanpy_macros2.xml	Thu Mar 12 09:12:26 2020 +0000
@@ -0,0 +1,142 @@
+<macros>
+  <token name="@TOOL_VERSION@">1.4.3</token>
+  <token name="@HELP@">More information can be found at https://scanpy.readthedocs.io</token>
+  <token name="@PROFILE@">18.01</token>
+  <token name="@VERSION_HISTORY@"><![CDATA[
+**Version history**
+
+1.4.3+galaxy10: Update to scanpy-scripts 0.2.10 (running scanpy ==1.4.3) to address bugfixes in run-pca.
+
+1.4.3+galaxy10: Update to scanpy-scripts 0.2.9 (running scanpy ==1.4.3) to address bugfixes in find-variable-genes.
+
+1.4.3+galaxy10: Use profile 18.01 for modules.
+
+1.4.3+galaxy6: Update to scanpy-scripts 0.2.8 (running scanpy ==1.4.3) and wider compatibility with other Galaxy modules. Bug fixes in filtering and plotting improvements.
+
+1.4.3+galaxy0: Update to scanpy-scripts 0.2.5 (running scanpy ==1.4.3).
+
+1.4.2+galaxy0: Update to scanpy-scripts 0.2.4 (requires scanpy >=1.4.2).
+
+1.3.2+galaxy1: Normalise-data and filter-genes: Exposes ability to output 10x files.
+
+1.3.2+galaxy0: Initial contribution. Ni Huang and Pablo Moreno, Expression Atlas team https://www.ebi.ac.uk/gxa/home  at
+EMBL-EBI https://www.ebi.ac.uk/ and Teichmann Lab at Wellcome Sanger Institute.
+    ]]></token>
+  <token name="@INPUT_OPTS@">
+    --input-format '${input_format}' input.h5
+  </token>
+  <token name="@OUTPUT_OPTS@">
+#if str($output_format).startswith('anndata')
+    --show-obj stdout --output-format anndata output.h5
+#else
+    --show-obj stdout --output-format loom output.h5
+#end if
+  </token>
+  <token name="@PLOT_OPTS@">
+#if $fig_title
+    --title '${fig_title}'
+#end if
+    --fig-size '${fig_size}'
+    --fig-dpi ${fig_dpi}
+    --fig-fontsize ${fig_fontsize}
+    ${fig_frame}
+    ./output.png
+  </token>
+  <token name="@EXPORT_MTX_OPTS@">${export_mtx}</token>
+
+  <xml name="requirements">
+    <requirements>
+      <requirement type="package" version="0.2.10">scanpy-scripts</requirement>
+      <yield/>
+    </requirements>
+  </xml>
+
+  <xml name="citations">
+    <citations>
+      <yield />
+      <citation type="doi">10.1186/s13059-017-1382-0</citation>
+      <citation type="bibtex">
+	@misc{githubscanpy-scripts,
+	author = {Ni Huang, EBI Gene Expression Team},
+	year = {2018},
+	title = {Scanpy-scripts: command line interface for Scanpy},
+	publisher = {GitHub},
+	journal = {GitHub repository},
+	url = {https://github.com/ebi-gene-expression-group/scanpy-scripts},
+      }</citation>
+    </citations>
+  </xml>
+
+  <xml name="input_object_params">
+    <param name="input_obj_file" argument="input-object-file" type="data" format="h5,h5ad" label="Input object in AnnData/Loom format"/>
+    <param name="input_format" argument="--input-format" type="select" label="Format of input object">
+      <option value="anndata" selected="true">AnnData format hdf5</option>
+      <option value="loom">Loom format hdf5</option>
+    </param>
+  </xml>
+
+  <xml name="output_object_params">
+    <param name="output_format" argument="--output-format" type="select" label="Format of output object">
+      <option value="anndata_h5ad" selected="true">AnnData format</option>
+      <option value="anndata">AnnData format (h5 for older versions)</option>
+      <option value="loom">Loom format</option>
+      <option value="loom_legacy">Loom format (h5 for older versions)</option>
+    </param>
+  </xml>
+
+  <xml name="output_object_params_no_loom">
+    <param name="output_format" argument="--output-format" type="select" label="Format of output object">
+      <option value="anndata_h5ad" selected="true">AnnData format</option>
+      <option value="anndata">AnnData format (h5 for older versions)</option>
+    </param>
+  </xml>
+
+  <xml name="output_data_obj_no_loom" token_description="operation">
+    <data name="output_h5ad" format="h5ad" from_work_dir="output.h5" label="${tool.name} on ${on_string}: @DESCRIPTION@ AnnData">
+      <filter>output_format == 'anndata_h5ad'</filter>
+    </data>
+    <data name="output_h5" format="h5" from_work_dir="output.h5" label="${tool.name} on ${on_string}: @DESCRIPTION@ AnnData">
+      <filter>output_format == 'anndata'</filter>
+    </data>
+  </xml>
+
+  <xml name="output_data_obj" token_description="operation">
+    <data name="output_h5ad" format="h5ad" from_work_dir="output.h5" label="${tool.name} on ${on_string}: @DESCRIPTION@ AnnData">
+      <filter>output_format == 'anndata_h5ad'</filter>
+    </data>
+    <data name="output_h5" format="h5" from_work_dir="output.h5" label="${tool.name} on ${on_string}: @DESCRIPTION@ AnnData">
+      <filter>output_format == 'anndata'</filter>
+    </data>
+    <data name="output_loom_legacy" format="h5" from_work_dir="output.h5" label="${tool.name} on ${on_string}: @DESCRIPTION@ Loom">
+      <filter>output_format == 'loom_legacy'</filter>
+    </data>
+    <data name="output_loom" format="loom" from_work_dir="output.h5" label="${tool.name} on ${on_string}: @DESCRIPTION@ Loom">
+      <filter>output_format == 'loom'</filter>
+    </data>
+  </xml>
+
+  <xml name="output_plot_params">
+    <param name="fig_title" argument="--title" type="text" label="Figure title"/>
+    <param name="fig_size" argument="--fig-size" type="text" value="4,4" label="Figure size as 'width,height', e.g, '7,7'"/>
+    <param name="fig_dpi" argument="--fig-dpi" type="integer" min="1" value="80" label="Figure dpi"/>
+    <param name="fig_fontsize" argument="--fig-fontsize" type="integer" min="0" value="10" label="Figure font size"/>
+    <param name="fig_frame" type="boolean" truevalue="--frameon" falsevalue="--frameoff" checked="false"
+           label="Show plot frame"/>
+  </xml>
+
+  <xml name="export_mtx_params">
+    <param name="export_mtx" argument="--export-mtx" type="boolean" truevalue="--export-mtx ./" falsevalue="" checked="false" label="Save to 10x mtx format" help="If enabled, it will generate in addition to the main output in Loom or AnnData an export in 10x format."/>
+  </xml>
+
+  <xml name="export_mtx_outputs">
+    <data name="matrix_10x" format="txt" from_work_dir="matrix.mtx" label="${tool.name} on ${on_string}: 10x matrix">
+      <filter>export_mtx</filter>
+    </data>
+    <data name="genes_10x" format="tsv" from_work_dir="genes.tsv" label="${tool.name} on ${on_string}: 10x genes">
+      <filter>export_mtx</filter>
+    </data>
+    <data name="barcodes_10x" format="tsv" from_work_dir="barcodes.tsv" label="${tool.name} on ${on_string}: 10x barcodes">
+      <filter>export_mtx</filter>
+    </data>
+  </xml>
+</macros>