Mercurial > repos > ebi-gxa > scanpy_filter_cells
changeset 1:4c806f3f094c draft
"planemo upload for repository https://github.com/ebi-gene-expression-group/container-galaxy-sc-tertiary/tree/develop/tools/tertiary-analysis/scanpy commit ba0c88ab1b077a0b1c60c8d3c529e72ca6946226"
author | ebi-gxa |
---|---|
date | Thu, 12 Mar 2020 09:07:07 +0000 |
parents | cad451be0a2a |
children | fd03785d0935 |
files | scanpy-filter-cells.xml scanpy_macros.xml scanpy_macros2.xml |
diffstat | 3 files changed, 219 insertions(+), 29 deletions(-) [+] |
line wrap: on
line diff
--- a/scanpy-filter-cells.xml Wed Mar 13 11:51:56 2019 -0400 +++ b/scanpy-filter-cells.xml Thu Mar 12 09:07:07 2020 +0000 @@ -1,46 +1,74 @@ <?xml version="1.0" encoding="utf-8"?> -<tool id="scanpy_filter_cells" name="Scanpy FilterCells" version="@TOOL_VERSION@+galaxy0"> +<tool id="scanpy_filter_cells" name="Scanpy FilterCells" version="@TOOL_VERSION@+galaxy10" profile="@PROFILE@" > <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}' +PYTHONIOENCODING=utf-8 scanpy-filter-cells +#if $gene_name + --gene-name '${gene_name}' +#end if +#if $parameters +#for $p in $parameters + #set $min = $p.min + #set $max = $p.max + #if $p.name.startswith('pct_') + #set $min = float($min) / 100 + #set $max = float($max) / 100 #end if - #if $subset - -s '${subset}' - #end if + --param 'c:$p.name' $min $max +#end for +#end if +#if $categories + #set cats = ' '.join(["--category 'c:{name}' '{negate}{values}'".format(**$c) for $c in $categories]) + ${cats} +#end if +#if $subsets + #set subs = ' '.join(["--subset 'c:{name}' '{subset}'".format(**$s) for $s in $subsets]) + ${subs} +#end if + @INPUT_OPTS@ + @OUTPUT_OPTS@ + @EXPORT_MTX_OPTS@ ]]></command> <inputs> <expand macro="input_object_params"/> <expand macro="output_object_params"/> - <repeat name="parameters" title="Parameters used to filter cells" min="1"> + + <param name="gene_name" type="text" optional="true" label="Name of the column in `anndata.var` that contains gene name" + help="Used for flagging mitochondria genes (starting with 'MT-'). Leave empty if gene table already has a boolean column called 'mito' that flags MT genes"/> + + <repeat name="parameters" title="Parameters to select cells to keep" 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="pct_counts_mito">pct_counts_mito (only usable if MT genes are flagged or has been pre-calculated)</option> </param> - <param name="min" type="float" value="0" min="0" label="Min value"/> - <param name="max" type="float" value="1e9" label="Max value"/> + <param name="min" type="float" value="0" min="0" label="Min value" help="Cells with value below min will be discarded."/> + <param name="max" type="float" value="1e9" label="Max value" help="Cells with value above max will be discarded."/> </repeat> - <param name="subset" argument="--subset-list" type="data" format="tsv" optional="true" label="List of barcodes"/> + + <repeat name="categories" title="Categories to select cells to keep (unless negate is checked)" 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 list of categories" help="Cells with these values in this categorical variable will be kept."/> + <param name="negate" type="boolean" truevalue="!" falsevalue="" checked="false" label="Apply as negative filter" help="If enabled, specified categories will be removed rather than retained."/> + </repeat> + + <repeat name="subsets" title="Subsets to select cells to keep" 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 to keep" help="A one-column headerless text file is required"/> + </repeat> + <param name="force_recalc" label="Force recalculation of QC vars" type="boolean" truevalue="--force-recalc" falsevalue="" help="If set, it will recalculate pcts and other existing QC vars, overwriting existing ones."/> + <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="output_data_obj" description="Filtered cells"/> + <expand macro="export_mtx_outputs"/> </outputs> <tests> @@ -58,18 +86,19 @@ <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@
--- a/scanpy_macros.xml Wed Mar 13 11:51:56 2019 -0400 +++ b/scanpy_macros.xml Thu Mar 12 09:07:07 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 ${on_string}: 10x barcodes"> + <filter>export_mtx</filter> + </data> + </xml> </macros>
--- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/scanpy_macros2.xml Thu Mar 12 09:07:07 2020 +0000 @@ -0,0 +1,142 @@ +<macros> + <token name="@TOOL_VERSION@">1.4.3</token> + <token name="@HELP@">More information can be found at https://scanpy.readthedocs.io</token> + <token name="@PROFILE@">18.01</token> + <token name="@VERSION_HISTORY@"><![CDATA[ +**Version history** + +1.4.3+galaxy10: Update to scanpy-scripts 0.2.10 (running scanpy ==1.4.3) to address bugfixes in run-pca. + +1.4.3+galaxy10: Update to scanpy-scripts 0.2.9 (running scanpy ==1.4.3) to address bugfixes in find-variable-genes. + +1.4.3+galaxy10: Use profile 18.01 for modules. + +1.4.3+galaxy6: Update to scanpy-scripts 0.2.8 (running scanpy ==1.4.3) and wider compatibility with other Galaxy modules. Bug fixes in filtering and plotting improvements. + +1.4.3+galaxy0: Update to scanpy-scripts 0.2.5 (running scanpy ==1.4.3). + +1.4.2+galaxy0: Update to scanpy-scripts 0.2.4 (requires scanpy >=1.4.2). + +1.3.2+galaxy1: Normalise-data and filter-genes: Exposes ability to output 10x files. + +1.3.2+galaxy0: Initial contribution. Ni Huang and Pablo Moreno, Expression Atlas team https://www.ebi.ac.uk/gxa/home at +EMBL-EBI https://www.ebi.ac.uk/ and Teichmann Lab at Wellcome Sanger Institute. + ]]></token> + <token name="@INPUT_OPTS@"> + --input-format '${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>