Mercurial > repos > ebi-gxa > scanpy_filter_cells

--- a/scanpy-filter-cells.xml	Wed Apr 03 11:12:31 2019 -0400
+++ b/scanpy-filter-cells.xml	Mon Sep 16 08:19:52 2019 -0400
@@ -2,45 +2,63 @@
 <tool id="scanpy_filter_cells" name="Scanpy FilterCells" version="@TOOL_VERSION@+galaxy0">
   <description>based on counts and numbers of genes expressed</description>
   <macros>
-    <import>scanpy_macros.xml</import>
+    <import>scanpy_macros2.xml</import>
   </macros>
   <expand macro="requirements"/>
   <command detect_errors="exit_code"><![CDATA[
 ln -s '${input_obj_file}' input.h5 &&
-PYTHONIOENCODING=utf-8 scanpy-filter-cells.py
-    -i input.h5
-    -f '${input_format}'
-    -o output.h5
-    -F '${output_format}'
-    #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 $subset
-        -s '${subset}'
-    #end if
+PYTHONIOENCODING=utf-8 scanpy-filter-cells
+#if $gene_name
+    --gene-name '${gene_name}'
+#end if
+#if $parameters
+    #set pars = ' '.join(['--param {name} {min} {max}'.format(**$p) for $p in $parameters])
+    ${pars}
+#end if
+#if $categories
+    #set cats = ' '.join(['--category {name} {values}'.format(**$c) for $c in $categories])
+    ${cats}
+#end if
+#if $subsets
+    #set subs = ' '.join(['--subsets {name} {subset}'.format(**$s) for $c in $subsets])
+    ${subs}
+#end if
+    @INPUT_OPTS@
+    @OUTPUT_OPTS@
     ]]></command>

   <inputs>
     <expand macro="input_object_params"/>
     <expand macro="output_object_params"/>
+
+    <param name="gene_name" type="text" value="index" label="Name of the column in `anndata.var` that contains gene name"
+           help="This is used for flagging mitochondria genes (starting with 'MT-')"/>
+
     <repeat name="parameters" title="Parameters used to filter cells" min="1">
       <param name="name" type="text" value="n_genes" label="Name of parameter to filter on" help="for example n_genes or n_counts">
         <option value="n_genes">n_genes</option>
-        <option value="n_counts">n_counts</option>
+        <option value="cell:n_counts">n_counts</option>
+        <option value="pct_counts_mito">pct_counts_mito</option>
       </param>
       <param name="min" type="float" value="0" min="0" label="Min value"/>
       <param name="max" type="float" value="1e9" label="Max value"/>
     </repeat>
-    <param name="subset" argument="--subset-list" type="data" format="tsv" optional="true" label="List of barcodes"/>
+
+    <repeat name="categories" title="Categories used to filter cells" min="0">
+      <param name="name" type="text" value="" label="Name of the categorical variable to filter on"/>
+      <param name="values" type="text" value="" label="Comma-separated values"/>
+    </repeat>
+
+    <repeat name="subsets" title="Subsets used to filter cells" min="0">
+      <param name="name" type="text" value="" label="Name of the categorical variable to filter on"/>
+      <param name="subset" type="data" format="tabular" label="List of values"/>
+    </repeat>
+    <expand macro="export_mtx_params"/>
   </inputs>

   <outputs>
     <data name="output_h5" format="h5" from_work_dir="output.h5" label="${tool.name} on ${on_string}: Filtered cells"/>
+    <expand macro="export_mtx_outputs"/>
   </outputs>

   <tests>
@@ -54,22 +72,23 @@
         <param name="max" value="2500"/>
       </repeat>
       <repeat name="parameters">
-        <param name="name" value="n_counts"/>
+        <param name="name" value="cell:n_counts"/>
         <param name="min" value="0"/>
         <param name="max" value="1e9"/>
       </repeat>
-      <output name="output_h5" file="filter_cells.h5" ftype="h5" compare="sim_size"/>
+      <output name="output_h5" file="output.h5" ftype="h5" compare="sim_size"/>
     </test>
   </tests>

   <help><![CDATA[
-========================================================================================
-Filter cells outliers based on counts and numbers of genes expressed (`pp.filter_cells`)
-========================================================================================
+===================================================================
+Filter cells based on various QC metrics (`scanpy.pp.filter_cells`)
+===================================================================

-For instance, only keep cells with at least `min_counts` counts or
-`min_genes` genes expressed. This is to filter measurement outliers, i.e.,
-"unreliable" observations.
+For instance, only keep cells with at least `min_counts` and at most
+`max_counts` UMI and/or at least `min_genes` expressed genes and/or at most
+`max_mito_percent` mitocondria expression. This is to filter measurement
+outliers, i.e., "unreliable" observations.

 @HELP@
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/scanpy_macros2.xml	Mon Sep 16 08:19:52 2019 -0400
@@ -0,0 +1,94 @@
+<macros>
+  <token name="@TOOL_VERSION@">1.4.2</token>
+  <token name="@HELP@">More information can be found at https://scanpy.readthedocs.io</token>
+  <token name="@VERSION_HISTORY@"><![CDATA[
+**Version history**
+
+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 '${input_format}' input.h5
+  </token>
+  <token name="@OUTPUT_OPTS@">
+    --show-obj stdout --output-format '${output_format}' output.h5
+  </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="0.2.4.post4">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" label="Input object in hdf5 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" selected="true">AnnData format hdf5</option>
+      <option value="loom">Loom format hdf5</option>
+    </param>
+  </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 normalised data 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 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>