changeset 0:f6f189ce4ebc draft

planemo upload for repository https://github.com/ebi-gene-expression-group/container-galaxy-sc-tertiary/tree/develop/tools/tertiary-analysis/scanpy commit 9bf9a6e46a330890be932f60d1d996dd166426c4
author ebi-gxa
date Wed, 03 Apr 2019 11:10:51 -0400
parents
children 2e74fd7b5f45
files scanpy-run-tsne.xml scanpy_macros.xml
diffstat 2 files changed, 236 insertions(+), 0 deletions(-) [+]
line wrap: on
line diff
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/scanpy-run-tsne.xml	Wed Apr 03 11:10:51 2019 -0400
@@ -0,0 +1,127 @@
+<?xml version="1.0" encoding="utf-8"?>
+<tool id="scanpy_run_tsne" name="Scanpy RunTSNE" version="@TOOL_VERSION@+galaxy1">
+  <description>visualise cell clusters using tSNE</description>
+  <macros>
+    <import>scanpy_macros.xml</import>
+  </macros>
+  <expand macro="requirements"/>
+  <command detect_errors="exit_code"><![CDATA[
+ln -s '${input_obj_file}' input.h5 &&
+PYTHONIOENCODING=utf-8 scanpy-run-tsne.py
+    -i input.h5
+    -f '${input_format}'
+    -o output.h5
+    -F '${output_format}'
+    #if $embeddings
+        --output-embeddings-file embeddings.csv
+    #end if
+    #if $settings.default == "false"
+        #if $settings.perplexity_file
+          --perplexity \$( cat $settings.perplexity_file )
+        #else
+          --perplexity '${settings.perplexity}'
+        #end if
+        --early-exaggeration '${settings.early_exaggeration}'
+        --learning-rate '${settings.learning_rate}'
+        #if $settings.use_rep != "auto"
+            -r '${settings.use_rep}'
+        #end if
+        #if $settings.n_pc
+            -n '${settings.n_pc}'
+        #end if
+        #if not $settings.fast_tsne
+            --no-fast-tsne
+        #end if
+        #if $settings.n_job
+            --n-jobs '${settings.n_job}'
+        #end if
+        #if $settings.random_seed is not None
+            -s '${settings.random_seed}'
+        #end if
+    #end if
+
+@PLOT_OPTS@
+]]></command>
+
+  <inputs>
+    <expand macro="input_object_params"/>
+    <expand macro="output_object_params"/>
+    <param name="embeddings" type="boolean" checked="true" label="Output embeddings in csv format"/>
+
+    <conditional name="settings">
+      <param name="default" type="boolean" checked="true" label="Use programme defaults"/>
+      <when value="true"/>
+      <when value="false">
+        <param name="use_rep" argument="--use-rep" type="select" label="Use the indicated representation">
+          <option value="X_pca">X_pca, use PCs</option>
+          <option value="X">X, use normalised expression values</option>
+          <option value="auto" selected="true">Automatically chosen based on problem size</option>
+        </param>
+        <param name="perplexity" argument="--perplexity" type="float" value="30" label="The perplexity is related to the number of nearest neighbours, select a value between 5 and 50"/>
+        <param name="perplexity_file" argument="--perplexity" type="data" format="txt,tsv" label="The perplexity is related to the number of nearest neighbours" help="For use with the parameter iterator. Overrides the persplexity option above" optional="true"/>
+        <param name="early_exaggeration" argument="--early-exaggeration" type="float" value="12" label="Controls the tightness within and between clusters"/>
+        <param name="learning_rate" argument="--learning-rate" type="float" value="1000" label="Learning rate, should be between 100 and 1000"/>
+        <param name="fast_tsne" type="boolean" checked="true" label="Use multicoreTSNE"/>
+        <param name="n_job" argument="--n-jobs" type="integer" optional="true" label="The number of jobs"/>
+        <param name="n_pc" argument="--n-pcs" type="integer" optional="true" label="The number of PCs to use"/>
+        <param name="random_seed" argument="--random-seed" type="integer" value="0" label="Seed for random number generator"/>
+      </when>
+    </conditional>
+
+    <conditional name="do_plotting">
+      <param name="plot" type="boolean" checked="false" label="Make tSNE plot"/>
+      <when value="true">
+        <expand macro="output_plot_params"/>
+        <param name="color_by" argument="--color-by" type="text" value="louvain" label="Color by attributes, comma separated strings"/>
+      </when>
+      <when value="false"/>
+    </conditional>
+  </inputs>
+
+  <outputs>
+    <data name="output_h5" format="h5" from_work_dir="output.h5" label="${tool.name} on ${on_string}: tSNE object"/>
+    <data name="output_png" format="png" from_work_dir="output.png" label="${tool.name} on ${on_string}: tSNE plot">
+      <filter>do_plotting['plot']</filter>
+    </data>
+    <data name="output_embed" format="csv" from_work_dir="embeddings.csv" label="${tool.name} on ${on_string}: tSNE embeddings">
+      <filter>embeddings</filter>
+    </data>
+  </outputs>
+
+  <tests>
+    <test>
+      <param name="input_obj_file" value="find_cluster.h5"/>
+      <param name="input_format" value="anndata"/>
+      <param name="output_format" value="anndata"/>
+      <param name="default" value="false"/>
+      <param name="embeddings" value="true"/>
+      <param name="random_seed" value="0"/>
+      <param name="plot" value="true"/>
+      <param name="color_by" value="louvain"/>
+      <output name="output_h5" file="run_tsne.h5" ftype="h5" compare="sim_size"/>
+      <output name="output_png" file="run_tsne.png" ftype="png" compare="sim_size"/>
+      <output name="output_embed" file="run_tsne.embeddings.csv" ftype="csv" compare="sim_size">
+        <assert_contents>
+          <has_n_columns n="2" sep=","/>
+        </assert_contents>
+      </output>
+    </test>
+  </tests>
+
+  <help><![CDATA[
+==================================================================
+t-distributed stochastic neighborhood embedding (tSNE) (`tl.tsne`)
+==================================================================
+
+t-distributed stochastic neighborhood embedding (tSNE) (Maaten et al, 2008) has been
+proposed for visualizating single-cell data by (Amir et al, 2013). Here, by default,
+we use the implementation of *scikit-learn* (Pedregosa et al, 2011).
+
+It yields `X_tsne`, tSNE coordinates of data.
+
+@HELP@
+
+@VERSION_HISTORY@
+]]></help>
+  <expand macro="citations"/>
+</tool>
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/scanpy_macros.xml	Wed Apr 03 11:10:51 2019 -0400
@@ -0,0 +1,109 @@
+<macros>
+  <token name="@TOOL_VERSION@">1.3.2</token>
+  <token name="@HELP@">More information can be found at https://scanpy.readthedocs.io</token>
+  <token name="@PLOT_OPTS@">
+#if $do_plotting.plot
+                  -P output.png
+                  --projectio $do_plotting.projection
+                  --components $do_plotting.components
+    #if $do_plotting.color_by
+                  --color-by $do_plotting.color_by
+    #end if
+    #if $do_plotting.groups
+                  --group $do_plotting.groups
+    #end if
+    #if $do_plotting.use_raw
+                  --use-raw
+    #end if
+    #if $do_plotting.palette
+                  --palette $do_plotting.palette
+    #end if
+    #if $do_plotting.edges
+                  --edges
+    #end if
+    #if $do_plotting.arrows
+                  --arrows
+    #end if
+    #if not $do_plotting.sort_order
+                  --no-sort-order
+    #end if
+    #if $do_plotting.frameoff
+                  --frameoff
+    #end if
+#end if
+  </token>
+  <xml name="requirements">
+    <requirements>
+      <requirement type="package" version="0.0.5">scanpy-scripts</requirement>
+      <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>
+  <xml name="citations">
+    <citations>
+      <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>
+      <yield />
+    </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, current support is incomplete</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, current support is defective</option>
+    </param>
+  </xml>
+  <xml name="output_plot_params">
+    <param name="color_by" argument="--color-by" type="text" value="n_genes" label="Color by attributes, comma separated strings"/>
+    <param name="groups" argument="--groups" type="text" optional="ture" label="Restrict plotting to named groups, comma separated strings"/>
+    <param name="projection" argument="--projection" type="select" label="Plot projection">
+      <option value="2d" selected="true">2D</option>
+      <option value="3d">3D</option>
+    </param>
+    <param name="components" argument="--components" type="text" value="1,2" label="Components to plot, comma separated integers"/>
+    <param name="palette" argument="--palette" type="text" optional="true" label="Palette"/>
+    <param name="use_raw" argument="--use-raw" type="boolean" checked="false" label="Use raw attributes if present"/>
+    <param name="edges" argument="--edges" type="boolean" checked="false" label="Show edges"/>
+    <param name="arrows" argument="--arrows" type="boolean" checked="false" label="Show arrows"/>
+    <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>