<tool id="episcanpy_preprocess" name="scATAC-seq Preprocessing" version="@TOOL_VERSION@+galaxy@VERSION_SUFFIX@" profile="@PROFILE@">
    <description>with EpiScanpy</description>
    <macros>
        <import>macros.xml</import>
    </macros>
    <expand macro="bio_tools"/>
    <expand macro="requirements"/>
    <expand macro="version_command"/>
    <command detect_errors="exit_code"><![CDATA[
@CMD@
      ]]></command>
    <configfiles>
        <configfile name="script_file"><![CDATA[
@CMD_imports@
@CMD_read_inputs@

#if $method.method == 'pp.binarize'
esc.pp.binarize(
    adata,
    copy=False)

#else if $method.method == 'pp.filter_cells'
esc.pp.filter_cells(
    adata,
    #if $method.filter.filter == 'min_counts'
    min_counts=$method.filter.min_counts,
    #else if $method.filter.filter == 'max_counts'
    max_counts=$method.filter.max_counts,
    #else if $method.filter.filter == 'min_features'
    min_features=$method.filter.min_features,
    #else if $method.filter.filter == 'max_features'
    max_features=$method.filter.max_features,
    #end if
    copy=False)

#else if $method.method == 'pp.filter_features'
esc.pp.filter_features(
    adata,
    #if $method.filter.filter == 'min_counts'
    min_counts=$method.filter.min_counts,
    #else if $method.filter.filter == 'max_counts'
    max_counts=$method.filter.max_counts,
    #else if $method.filter.filter == 'min_cells'
    min_cells=$method.filter.min_cells,
    #else if $method.filter.filter == 'max_cells'
    max_cells=$method.filter.max_cells,
    #end if
    copy=False)

#else if $method.method == 'nb_feat_log'
adata.obs['log_nb_features'] = [np.log10(x) for x in adata.obs['nb_features']]

#else if $method.method == 'pp.coverage_cells'
esc.pp.coverage_cells(
    adata,
    binary=$method.binary,
    log=$method.log,
    #if $method.threshold
    threshold=$method.threshold,
    #end if
    bins=$method.bins,
    save='plot.png'
    )
#else if $method.method == 'pp.coverage_features'
esc.pp.coverage_features(
    adata,
    binary=$method.binary,
    log=$method.log,
    #if $method.threshold
    threshold=$method.threshold,
    #end if
    bins=$method.bins,
    save='plot.png'
    )

#else if $method.method == 'pp.select_var_feature'
esc.pp.select_var_feature(
    adata,
    min_score=$method.min_score,
    show=False,
    #if $method.nb_features
    nb_features=$method.nb_features
    #end if
)

#else if $method.method == 'pp.cal_var'
esc.pp.cal_var(
    adata,
    show=True,
    color=['b', 'r'],
    save='plot.png'
)

#else if $method.method == 'pp.variability_features'
esc.pp.variability_features(
    adata,
    min_score=$method.min_score,
    nb_features=$method.nb_features,
    #if $method.log_mode
    log='$method.log_mode',
    #end if
    save='plot.png'
)

#else if $method.method == 'tl.find_genes'
esc.tl.find_genes(
    adata,
    gtf_file='$method.find_genes_gtf_file',
    key_added='$method.find_genes_key_added',
    upstream=$method.find_genes_upstream,
    feature_type='$method.find_genes_feature_type',
    annotation='$method.find_genes_annotation',
    raw=$method.find_genes_raw)

#end if
@CMD_anndata_write_outputs@
]]></configfile>
    </configfiles>
    <inputs>
        <expand macro="inputs_anndata"/>
        <conditional name="method">
            <param argument="method" type="select" label="Method used for filtering">
                <option value="pp.binarize">Binarize count matrix, using 'pp.binarize'</option>
                <option value="pp.filter_cells">Filter cell outliers based on counts and numbers of features expressed, using 'pp.filter_cells'</option>
                <option value="pp.filter_features">Filter features based on counts and numbers of features expressed, using 'pp.filter_features'</option>
                <option value="nb_feat_log">Compute log10 of nb_features</option>
                <option value="pp.coverage_cells">Coverage cells: Histogram of the number of open features (in the case of ATAC-seq data) per cell, using 'pp.coverage_cells'</option>
                <option value="pp.coverage_features">Coverage features: Distribution of the feature commoness in cells, using 'pp.coverage_features'</option>
                <option value="pp.select_var_feature">Select the most variable features, 'pp.select_var_feature'</option>
                <option value="pp.cal_var">Show distribution plots of cells sharing features and variability score 'pp.cal_var'</option>
                <option value="pp.variability_features">Computes variability score to rank the most variable features across all cells, using 'pp.variability_features'</option>
                <option value="tl.find_genes">Embedding: Find genes and add annotations, using 'pp.find_genes'</option>
            </param>
            <when value="pp.binarize" />
            <when value="pp.filter_cells">
                <conditional name="filter">
                    <param argument="filter" type="select" label="Filter" help="Filter mode">
                        <option value="min_counts">Minimum number of counts</option>
                        <option value="max_counts">Maximum number of counts</option>
                        <option value="min_features">Minimum number of features expressed</option>
                        <option value="max_features">Maximum number of features expressed</option>
                    </param>
                    <when value="min_counts">
                        <param argument="min_counts" type="integer" min="0" value="" label="Minimum counts" help="Minimum number of counts required for a cell to pass filtering"/>
                    </when>
                    <when value="max_counts">
                        <param argument="max_counts" type="integer" min="0" value="" label="Maximum counts" help="Maximum number of counts required for a cell to pass filtering"/>
                    </when>
                    <when value="min_features">
                        <param argument="min_features" type="integer" min="0" value="" label="Minimum  features" help="Minimum number of features expressed required for a cell to pass filtering"/>
                    </when>    
                    <when value="max_features">
                        <param argument="max_features" type="integer" min="0" value="" label="Maximum features" help="Maximum number of features expressed required for a cell to pass filtering"/>
                    </when>
                </conditional>
            </when>
            <when value="pp.filter_features">
                <conditional name="filter">
                    <param argument="filter" type="select" label="Filter">
                        <option value="min_counts">Minimum number of counts</option>
                        <option value="max_counts">Maximum number of counts</option>
                        <option value="min_cells">Minimum number of cells expressed</option>
                        <option value="max_cells">Maximum number of cells expressed</option>
                    </param>
                    <when value="min_counts">
                        <param argument="min_counts" type="integer" min="0" value="" label="Minimum number of counts required for a gene to pass filtering"/>
                    </when>
                    <when value="max_counts">
                        <param argument="max_counts" type="integer" min="0" value="" label="Maximum number of counts required for a gene to pass filtering"/>
                    </when>
                    <when value="min_cells">
                        <param argument="min_cells" type="integer" min="0" value="" label="Minimum number of cells expressed required for a gene to pass filtering"/>
                    </when>    
                    <when value="max_cells">
                        <param argument="max_cells" type="integer" min="0" value="" label="Maximum number of cells expressed required for a gene to pass filtering"/>
                    </when>
                </conditional>
            </when>
            <when value="nb_feat_log" />
            <when value="pp.coverage_cells">
                <expand macro="coverage_params" />
            </when>
            <when value="pp.coverage_features">
                <expand macro="coverage_params" />
            </when>
            <when value="pp.select_var_feature">
                <param argument="min_score" type="float" min="0" max="1" value="0.5" label="Min score" help="Minimum threshold variability score to retain features" />
                <param argument="nb_features" type="integer" min="0" value="" optional="True" label="Number of features" help="Default value is None, if specify it will select a the top most 
                    variable features. If this parameter is larger than the total number of feature, it filters based on the min_score argument." />
            </when>
            <when value="pp.cal_var" />
            <when value="pp.variability_features">
                <param name="min_score" type="float" min="0" max="1" value="0.5" label="Minimum score value"/>
                <param name="nb_features" type="integer" min="0" value="" label="Number of features"/>
                <param name="log_mode" type="select" optional="True" label="Log" help="Log mode">
                    <option value="log2">Log2</option>
                    <option value="log10">Log10</option>
                </param>
            </when>
            <when value="tl.find_genes">
                <param name="find_genes_gtf_file" type="data" format="gtf" label="Annotation GTF file" help="(gtf_file)"/>
                <param name="find_genes_key_added" value="transcript_annotation" type="text" label="Key added" help="(key_added)">
                    <sanitizer invalid_char="">
                        <valid initial="string.letters,string.digits">
                            <add value="_" />
                            <add value="-" />
                            <add value="." />
                            <add value=" " />
                            <add value="," />
                        </valid>
                    </sanitizer>
                    <validator type="regex">[0-9a-zA-Z_., -]+</validator>
                </param>
                <param name="find_genes_upstream" value="2000" min="0" type="integer" label="Upstream" help="(upstream)"/>
                <param name="find_genes_feature_type" value="transcript" type="text" label="Feature type" help="(feature_type)">
                    <sanitizer invalid_char="">
                        <valid initial="string.letters,string.digits">
                            <add value="_" />
                            <add value="-" />
                            <add value="." />
                            <add value=" " />
                            <add value="," />
                        </valid>
                    </sanitizer>
                    <validator type="regex">[0-9a-zA-Z_., -]+</validator>
                </param>
                <param name="find_genes_annotation" value="HAVANA" type="text" label="Annotation" help="(annotation)">
                    <sanitizer invalid_char="">
                        <valid initial="string.letters,string.digits">
                            <add value="_" />
                            <add value="-" />
                            <add value="." />
                            <add value=" " />
                            <add value="," />
                        </valid>
                    </sanitizer>
                    <validator type="regex">[0-9a-zA-Z_., -]+</validator>
                </param>
                <param name="find_genes_raw" type="select" label="Raw?" help="(raw)">
                    <option value="True">True</option>
                    <option value="False" selected="true">False</option>
                </param>
            </when>
        </conditional>
        <expand macro="inputs_common_advanced"/>
    </inputs>
    <outputs>
        <expand macro="anndata_outputs"/>
        <data name="out_png" format="png" from_work_dir="plot.png" label="${tool.name} (${method.method}) plot on ${on_string}">
            <filter>method['method'] != 'pp.binarize' and method['method'] != 'pp.filter_cells' and method['method'] != 'pp.filter_features' and method['method'] != 'nb_feat_log' and method['method'] != 'pp.select_var_feature' and method['method'] !='tl.find_genes'</filter>
        </data>
    </outputs>
    <tests>
        <test expect_num_outputs="1">
            <!-- pp.binarize -->
            <param name="adata" value="krumsiek11.h5ad" />
            <conditional name="method">
                <param name="method" value="pp.binarize"/>
            </conditional>
            <output name="anndata_out" file="krumsiek11.pp.binarize.h5ad" ftype="h5ad" compare="sim_size"/>
        </test>
        <test expect_num_outputs="2">
            <!-- pp.filter_cells -->
            <param name="adata" value="krumsiek11.h5ad" />
            <conditional name="method">
                <param name="method" value="pp.filter_cells"/>
                <conditional name="filter">
                    <param name="filter" value="min_features"/>
                    <param name="min_features" value="10"/>
                </conditional>
            </conditional>
            <section name="advanced_common">
                <param name="show_log" value="true" />
            </section>
            <assert_stdout>
                <has_text_matching expression="395 × 11"/>
            </assert_stdout>
            <output name="hidden_output">
                <assert_contents>
                    <has_text_matching expression="esc.pp.filter_cells"/>
                    <has_text_matching expression="min_features=10"/>
                </assert_contents>
            </output>
            <output name="anndata_out" file="krumsiek11.pp.filter_cells.min_features.h5ad" ftype="h5ad" compare="sim_size"/>
        </test>
        <test expect_num_outputs="2">
            <!-- pp.filter_features -->
            <param name="adata" value="krumsiek11.h5ad" />
            <conditional name="method">
                <param name="method" value="pp.filter_features"/>
                <conditional name="filter">
                    <param name="filter" value="min_cells"/>
                    <param name="min_cells" value="600"/>
                </conditional>
            </conditional>
            <section name="advanced_common">
                <param name="show_log" value="true" />
            </section>
            <assert_stdout>
                <has_text_matching expression="640 × 2"/>
            </assert_stdout>
            <output name="hidden_output">
                <assert_contents>
                    <has_text_matching expression="esc.pp.filter_features"/>
                    <has_text_matching expression="min_cells=600"/>
                </assert_contents>
            </output>
            <output name="anndata_out" file="krumsiek11.pp.filter_features.min_cells.h5ad" ftype="h5ad" compare="sim_size"/>
        </test>
        <test expect_num_outputs="1">
            <!-- nb_feat_log -->
            <param name="adata" value="krumsiek11.pp.filter_cells.min_features.h5ad" />
            <conditional name="method">
                <param name="method" value="nb_feat_log"/>
            </conditional>
            <assert_stdout>
                <has_text_matching expression="log_nb_features"/>
                <has_text_matching expression="nb_features"/>
            </assert_stdout>
        </test>
        <test  expect_num_outputs="2">
            <!-- pp.select_var_feature -->
            <param name="adata" value="krumsiek11.h5ad" />
            <conditional name="method">
                <param name="method" value="pp.select_var_feature"/>
                <param name="min_score" value="0.6"/>
                <param name="nb_features" value="10"/>
            </conditional>
            <section name="advanced_common">
                <param name="show_log" value="true" />
            </section>
            <assert_stdout>
                <has_text_matching expression="prop_shared_cells"/>
                <has_text_matching expression="variability_score"/>
            </assert_stdout>
            <output name="hidden_output">
                <assert_contents>
                    <has_text_matching expression="esc.pp.select_var_feature"/>
                    <has_text_matching expression="adata"/>
                    <has_text_matching expression="min_score=0.6"/>
                    <has_text_matching expression="nb_features=10"/>
                </assert_contents>
            </output>
            <output name="anndata_out" file="krumsiek11.pp.select_var_feature.h5ad" ftype="h5ad" compare="sim_size"/>
        </test>
        <test expect_num_outputs="3">
            <!-- pp.cal_var -->
            <param name="adata" value="krumsiek11.h5ad" />
            <conditional name="method">
                <param name="method" value="pp.cal_var"/>
            </conditional>
            <section name="advanced_common">
                <param name="show_log" value="true" />
            </section>
            <output name="hidden_output">
                <assert_contents>
                    <has_text_matching expression="esc.pp.cal_var"/>
                    <has_text_matching expression="adata"/>
                    <has_text_matching expression="plot.png"/>
                </assert_contents>
            </output>
            <output name="anndata_out" file="krumsiek11.pp.cal_var.h5ad" ftype="h5ad" compare="sim_size"/>
            <output name="out_png" file="krumsiek11.pp.cal_var.png" ftype="png" compare="sim_size"/>
        </test>
        <test expect_num_outputs="3">
            <!-- pp.coverage_cells -->
            <param name="adata" value="krumsiek11.h5ad" />
            <conditional name="method">
                <param name="method" value="pp.coverage_cells"/>
                <param name="threshold" value="3.0" />
                <param name="binary" value="True" />
            </conditional>
            <section name="advanced_common">
                <param name="show_log" value="true" />
            </section>
            <output name="hidden_output">
                <assert_contents>
                    <has_text_matching expression="esc.pp.coverage_cells"/>
                    <has_text_matching expression="adata"/>
                    <has_text_matching expression="threshold=3.0"/>
                </assert_contents>
            </output>
            <output name="anndata_out" file="krumsiek11.pp.coverage_cells.h5ad" ftype="h5ad" compare="sim_size"/>
            <output name="out_png" file="krumsiek11.pp.coverage_cells.png" ftype="png" compare="sim_size"/>
        </test>
        <test expect_num_outputs="3">
            <!-- pp.coverage_features -->
            <param name="adata" value="krumsiek11.h5ad" />
            <conditional name="method">
                <param name="method" value="pp.coverage_features"/>
                <param name="threshold" value="100" />
                <param name="binary" value="True" />
            </conditional>
            <section name="advanced_common">
                <param name="show_log" value="true" />
            </section>
            <output name="hidden_output">
                <assert_contents>
                    <has_text_matching expression="esc.pp.coverage_features"/>
                    <has_text_matching expression="adata"/>
                    <has_text_matching expression="threshold=100"/>
                </assert_contents>
            </output>
            <output name="anndata_out" file="krumsiek11.pp.coverage_features.h5ad" ftype="h5ad" compare="sim_size"/>
            <output name="out_png" file="krumsiek11.pp.coverage_features.png" ftype="png" compare="sim_size"/>
        </test>
        <test expect_num_outputs="3">
            <!-- pp.coverage_features -->
            <param name="adata" value="krumsiek11.h5ad" />
            <conditional name="method">
                <param name="method" value="pp.variability_features"/>
                <param name="min_score" value="0.75" />
                <param name="nb_features" value="8" />
                <param name="log" value="log10" />
            </conditional>
            <section name="advanced_common">
                <param name="show_log" value="true" />
            </section>
            <output name="hidden_output">
                <assert_contents>
                    <has_text_matching expression="esc.pp.variability_features"/>
                    <has_text_matching expression="adata"/>
                    <has_text_matching expression="min_score=0.75"/>
                    <has_text_matching expression="nb_features=8"/>
                </assert_contents>
            </output>
            <output name="anndata_out" file="krumsiek11.pp.variability_features.h5ad" ftype="h5ad" compare="sim_size"/>
            <output name="out_png" file="krumsiek11.pp.variability_features.png" ftype="png" compare="sim_size"/>
        </test>
       <test expect_num_outputs="2">
           <!-- tl.find_genes -->
           <param name="adata" value="chrY.h5ad" />
           <conditional name="method">
               <param name="method" value="tl.find_genes"/>
               <param name="find_genes_gtf_file" value="chrY.gtf"/>
               <param name="find_genes_key_added" value="transcript_annotation"/>
               <param name="find_genes_upstream" value="2000"/>
               <param name="find_genes_feature_type" value="transcript"/>
               <param name="find_genes_annotation" value="HAVANA"/>
               <param name="find_genes_raw" value="False"/>
           </conditional>
           <section name="advanced_common">
               <param name="show_log" value="true" />
           </section>
           <output name="anndata_out" file="chrY_with_transcript_annotation.h5ad" ftype="h5ad" compare="sim_size">
                <assert_contents>
                    <has_h5_keys keys="var" />
                </assert_contents>
            </output>
        </test>
    </tests>
    <help><![CDATA[

convert the count matrix into a binary matrix (`pp.binarize`)
============================================================================================
convert the count matrix into a binary matrix

More details on the `episcanpy documentation
<https://colomemaria.github.io/episcanpy_doc/api/episcanpy.api.pp.binarize.html>`__

Filter cells outliers based on counts and numbers of features expressed (`pp.filter_cells`)
============================================================================================
For instance, only keep cells with at least *min_counts* counts or *min_features* genes expressed. 
This is to filter measurement outliers, i.e. "unreliable" observations.

Only provide one of the optional parameters *min_counts*, *min_features*, *max_counts*, *max_features* per call.

More details on the `episcanpy documentation
<https://colomemaria.github.io/episcanpy_doc/api/episcanpy.api.pp.filter_cells.html>`__

Filter features based on number of cells or counts (`pp.filter_features`)
========================================================================================
Keep features that have at least *min_counts* counts or are expressed in at least *min_cells* cells or 
have at most *max_counts* counts or are expressed in at most *max_cells* cells.

Only provide one of the optional parameters *min_counts*, *min_cells*, *max_counts*, *max_cells* per call.

More details on the `episcanpy documentation
<https://colomemaria.github.io/episcanpy_doc/api/episcanpy.api.pp.filter_features.html>`__

Histogram of the number of open features (`pp.coverage_cells`)
========================================================================================
Histogram of the number of open features (in the case of ATAC-seq data) per cell.

More details on the `episcanpy documentation
<https://colomemaria.github.io/episcanpy_doc/api/episcanpy.api.pp.coverage_cells.html>`__

Distribution of the feature commoness in cells (`pp.coverage_features`)
========================================================================================
Display how often a feature is measured as open (for ATAC-seq). Distribution of the feature commoness in cells.

More details on the `episcanpy documentation
<https://colomemaria.github.io/episcanpy_doc/api/episcanpy.api.pp.coverage_features.html>`__

Selects the most variable features according to either a specified number of features or minimum variance score (`pp.select_var_feature`)
=========================================================================================================================================

This function computes a variability score to rank the most variable features across all cells. Then it selects the most variable features according to either a specified number of features (nb_features) or a minimum variance score (min_score).

More details on the `episcanpy documentation
<https://colomemaria.github.io/episcanpy_doc/api/episcanpy.api.pp.select_var_feature.html>`__

Distribution of cells sharing features and variability score (`pp.cal_var`)
=============================================================================
Show distribution plots of cells sharing features and variability score.

More details on the `episcanpy documentation
<https://colomemaria.github.io/episcanpy_doc/api/episcanpy.api.pp.cal_var.html>`__

Compute a variability score to rank the most variable features across all cells (`pp.variability_features`)
============================================================================================================
This function computes a variability score to rank the most variable features across all cells. Then it selects the most variable features according to either a specified number of features (nb_features) or a minimum variance score (min_score).

Find and add gene annotations (`tl.find_genes`)
========================================================================================
This function adds a gene annotation to an AnnData (h5ad) file from annotations file (.annotation.gtf).

More details on the `episcanpy documentation
<https://colomemaria.github.io/episcanpy_doc/api/episcanpy.api.pp.variability_features.html>`__
    ]]></help>
    <expand macro="citations"/>
</tool>