view evaluation.xml @ 0:0c1b0a3b19df draft default tip

planemo upload for repository https://github.com/galaxyproject/tools-iuc/tree/master/tools/spapros/ commit aed7fe13fa0ed09d77a31eeecaf3ec3fba7eed3b

<tool id="spapros_evaluation" name="Evaluation" version="@TOOL_VERSION@+galaxy@VERSION_SUFFIX@" profile="@profile@">
    <description>of marker genes with spapros</description>
    <macros>
        <import>macros.xml</import>
    </macros>
    <expand macro="requirements">
    </expand>
    <expand macro="version_command"/>
    <command detect_errors="exit_code"><![CDATA[
@CMD@
      ]]></command>
    <configfiles>
        <configfile name="script_file"><![CDATA[
@CMD_imports@
@CMD_read_inputs@

mpl.rcParams['figure.dpi'] = $figure_options.dpi
plt.rcParams["font.size"] = $figure_options.fontsize

header_probeset='infer'
#if $header_probeset == 'not_included':
header_probeset=None
#end if

## Probe set generation
probeset = pd.read_csv('$probeset', sep='\t', index_col=0, header=header_probeset)
probeset = [x for x in probeset.values.flatten().tolist() if not pd.isna(x)]

## Reference set generation
methods_pre="#echo ','.join(["%s" % $s for $i, $s in enumerate($methods_reference)])#"
methods=methods_pre.split(',')

reference_sets = sp.se.select_reference_probesets(
adata, 
n=$nreference,
#if $genes_key != '':
genes_key='$genes_key',
#else:
genes_key=None,
#end if
#if $obs_key != '':
obs_key='$obs_key',
#end if
#if $seeds != '':
seeds=[$seeds],
#end if
methods=methods,
verbosity=0
)

## Parameter setting

custom_params = {
'data': {
'name': '$probeset_name', 
'celltype_key': '$celltype_key'
}, 
'metrics': {
#if $method.method == 'plot_summary' or $method.method == 'plot_cluster_similarity':
'cluster_similarity': {
'ns': [$method.ns_start, $method.ns_end],
'AUC_borders': [#echo ",".join(["[%s, %s]" % ($s.auc_borders_start, $s.auc_borders_end) for $i, $s in enumerate($method.series_auc_borders)])#]
},
#end if
#if $method.method == 'plot_summary' or $method.method == 'plot_knn_overlap':
'knn_overlap': {
'ks': [#echo ",".join([ "%s" % $s.k for $i, $s in enumerate($method.knn) ])#]
},
#end if
#if $method.method == 'plot_confusion_matrix':
'forest_clfs': {
'ct_key': '$celltype_key',
#if $method.method == 'plot_summary':
'threshold': $method.forest_clfs_threshold
#end if
},
#end if
#if $method.method == 'plot_summary':
#if $method.select_marker_corr.use_marker_corr == 'True':
'marker_corr': {
'per_celltype': $method.select_marker_corr.per_celltype,
'per_marker': $method.select_marker_corr.per_marker,
#if $method.select_marker_corr.select_per_celltype_min_mean.use_per_celltype_min_mean == 'True'
'per_celltype_min_mean': $method.select_marker_corr.select_per_celltype_min_mean.per_celltype_min_mean,
#end if
#if $method.select_marker_corr.select_per_marker_min_mean.use_per_marker_min_mean == 'True'
'per_marker_min_mean': $method.select_marker_corr.select_per_marker_min_mean.per_marker_min_mean
#end if
},
#end if
#end if
#if $method.method == 'plot_marker_corr':
'marker_corr': {
'per_celltype': $method.per_celltype,
'per_marker': $method.per_marker,
#if $method.select_per_celltype_min_mean.use_per_celltype_min_mean == 'True'
'per_celltype_min_mean': $method.select_per_celltype_min_mean.per_celltype_min_mean,
#else:
'per_celltype_min_mean': None,
#end if
#if $method.select_per_marker_min_mean.use_per_marker_min_mean == 'True'
'per_marker_min_mean': $method.select_per_marker_min_mean.per_marker_min_mean
#else:
'per_marker_min_mean': None
#end if
},
#end if
#if $method.method == 'plot_summary':
'gene_corr': {
'threshold': $method.gene_corr_threshold
}
#end if
}
}

## Setup evaluator

#if $method.method == 'plot_summary' or $method.method == 'plot_marker_corr':

header_markerset='infer'
feature_marker=dict()

#if $method.method == 'plot_summary':
#if $method.select_marker_corr.header_markerset == 'not_included' :
header_markerset=None
#end if
feature_marker = {key: [v for v in list(value.values()) if pd.notna(v)] for key, value in pd.read_csv('$method.select_marker_corr.markerset', sep='\t', index_col=0, header=header_markerset).to_dict(orient='index').items()}
#end if
#if $method.method == 'plot_marker_corr':
#if $method.header_markerset == 'not_included':
header_markerset=None
#end if
feature_marker = {key: [v for v in list(value.values()) if pd.notna(v)] for key, value in pd.read_csv('$method.markerset', sep='\t', index_col=0, header=header_markerset).to_dict(orient='index').items()}
#end if

#end if

evaluator = sp.ev.ProbesetEvaluator(
adata, 
metrics_params=custom_params, 
scheme="full", 
verbosity=0,
#if $method.method == 'plot_summary' or $method.method == 'plot_marker_corr':
marker_list=feature_marker,
#end if
results_dir=None
)

## Evaluation reference set
for set_id, df in reference_sets.items():
    gene_set = df[df["selection"]].index.to_list()
    evaluator.evaluate_probeset(gene_set, set_id=set_id)

## Evaluation probe set
evaluator.evaluate_probeset(probeset, set_id='$probeset_name')

## Execution of method ##

#if $method.method == 'plot_summary':
evaluator.plot_summary(
@CMD_plot@
)
#end if

#if $method.method == 'plot_confusion_matrix':
evaluator.plot_confusion_matrix(
@CMD_plot@
)
#end if

#if $method.method == 'plot_coexpression':
evaluator.plot_coexpression(
@CMD_plot@
)
#end if

#if $method.method == 'plot_cluster_similarity':
evaluator.plot_cluster_similarity(
@CMD_plot@
)
#end if

#if $method.method == 'plot_knn_overlap':
evaluator.plot_knn_overlap(
@CMD_plot@
)
#end if

## plot_marker_corr does not allow for the show keyword
#if $method.method == 'plot_marker_corr':
evaluator.plot_marker_corr(
#if $method.set_ids != 'all' and $method.set_ids != '':
set_ids=[$method.set_ids],
#end if
save='plot.$format'
)
#end if

]]></configfile>
    </configfiles>
    <inputs>
        <expand macro="inputs_anndata"/>
        <expand macro="param_plot_format"/>
        <param name="probeset" type="data" format="tabular" label="Probeset tabular file with rows=conditions (e.g., celltypes) and column=features (e.g., genes)"/>
        <param name="header_probeset" type="select" optional="false" label="Header in the list of probes?">
            <option value="included">Header incldued</option>
            <option value="not_included">Header not included</option>
        </param>
        <param argument="nreference" type="integer" value="10" min="1" optional="false" label="Number of selected genes to generate a reference set"/> 
        <param name="methods_reference" type="select" multiple="True" optional="false" label="Select the methods to genereate a reference probeset">
            <option value="random" selected="true">Random</option>
            <option value="PCA" selected="false">PCA</option>
            <option value="DE" selected="false">Differential expressed features based on wilxocon rank genes test (DE)</option>
            <option value="HVG" selected="false">Highly variable features (HVG)</option>
        </param>
        <param name="genes_key" type="text" optional="true" label="adata.var key for subset of preselected genes to run the selections" help="This is typically highly_variable_genes. Leave empty to not subset genes."/>
        <param name="obs_key" type="text" optional="true" label="Column name of adata.obs for which marker scores are calculated" help="Only required for method DE."/>
        <param name="seeds" type="text" optional="true" label="List of random integer seeds (comma seperated, e.g., 123,999,22)" help="For each seed, one random gene set is selected if random in methods. Leave it empty to for one random set with a random seed."/>
        <param name="probeset_name" type="text" value="probeset" optional="false" label="Name for your probeset that should be displayed in plots"/>
        <param name="celltype_key" type="text" value="celltype" optional="false" label="Column name of adata.obs that represents groups of your cells (e.g., celltype)"/>
        <conditional name="method">
            <param argument="method" type="select" optional="false" label="Method of spapros">
                <option value="plot_summary">Plot statistic summary, using 'evaluator.plot_summary'</option>
                <option value="plot_confusion_matrix">Plot confusion matrix, using 'evaluator.plot_confusion_matrix'</option>
                <option value="plot_coexpression">Plot coexpression heatmap, using 'evaluator.plot_coexpression'</option>
                <option value="plot_cluster_similarity">Plot cluster similariy, using 'evaluator.plot_cluster_similarity'</option>
                <option value="plot_knn_overlap">Plot knn overlap, using 'evaluator.plot_knn_overlap'</option>
                <option value="plot_marker_corr">Plot featue marker correlation, using 'evaluator.plot_marker_corr'</option>
            </param>
            <when value="plot_summary">
                <expand macro="set_ids"/>
                <param name="forest_clfs_threshold" type="float" value="0.8" min="0.0" max="1.0" label="Assesses how many cell types (%) can be predicted with an accuracy of at least threshold" help="The metric gives an idea about how many cell types can be identified with high confidence with the given gene set."/>
                <param name="gene_corr_threshold" type="float" value="0.8" min="0.0" max="1.0" label="Percentage of features (e.g., genes) that have a maximum correlation of less than threshold with all other features" help="This metric gives an idea about how many features show unique expression profiles in the gene set."/>
                <param name="ns_start" type="integer" value="5" min="1" label="The minimum number of leiden clusters clusters" help="Clusterings are calculated with different leiden resolutions to genertate clusterings of n = ns_start to ns_end clusters."/>
                <param name="ns_end" type="integer" value="21" min="1" label="The maximum number of leiden clusters clusters" help="Clusterings are calculated with different leiden resolutions to genertate clusterings of n = ns_start to ns_end clusters."/>
                <repeat name="series_auc_borders" title="Calculates nmi AUCs over given borders">
                    <param name="auc_borders_start" type="integer" value="15" min="1" label="Calculates nmi over n ranges auc_borders_start to auc_borders_end" help="Defined border shouldn't exceed values in nmis."/>
                    <param name="auc_borders_end" type="integer" value="20" min="1" label="Calculates nmi over n ranges auc_borders_start to auc_borders_end" help="Defined border shouldn't exceed values in nmis."/>
                </repeat>
                <repeat name="knn" title="Calculate knn graphs for each k">
                    <param name="k" type="integer" value="10" min="1" label="Includes nearest neighbors for all k"/>
                </repeat>
                <conditional name="select_marker_corr">
                    <param name="use_marker_corr" type="select" label="Do you want to calculate the correlation between your probeset and marker features?">
                        <option value="False">No</option>
                        <option value="True">Yes</option>
                    </param>
                    <when value="True">
                        <expand macro="param_markerset"/>
                        <param name="per_celltype" type="boolean" truevalue="True" falsevalue="False" label="Wether to return columns with per cell type max correlations" checked="true"/>
                        <param name="per_marker" type="boolean" truevalue="True" falsevalue="False" label="Wether to return columns with per marker max correlations" checked="true"/>
                        <conditional name="select_per_celltype_min_mean">
                            <param name="use_per_celltype_min_mean" type="select" label="Add a column for correlation per cell type that only takes into accounts markers with mean expression > per_celltype_min_mean">
                                <option value="False">No</option>
                                <option value="True">Yes</option>
                            </param>
                            <when value="True">
                                <param name="per_celltype_min_mean" type="float" value="0.0" min="0.0" label="Set per_celltype_min_mean"/>
                            </when>
                            <when value="False"/>
                        </conditional>
                        <conditional name="select_per_marker_min_mean">
                            <param name="use_per_marker_min_mean" type="select" label="Add a column for correlation per cell type that only takes into accounts markers with mean expression > per_celltype_min_mean">
                                <option value="False">No</option>
                                <option value="True">Yes</option>
                            </param>
                            <when value="True">
                                <param name="per_marker_min_mean" type="float" value="0.0" min="0.0" label="Set per_marker_min_mean"/>
                            </when>
                            <when value="False"/>
                        </conditional>
                    </when>
                    <when value="False"/>
                </conditional>
            </when>
            <when value="plot_confusion_matrix">
                <expand macro="set_ids"/>
            </when>
            <when value="plot_coexpression">
                <expand macro="set_ids"/>
            </when>
            <when value="plot_cluster_similarity">
                <expand macro="set_ids"/>
                <param name="ns_start" type="integer" value="5" min="1" label="The minimum number of leiden clusters clusters" help="Clusterings are calculated with different leiden resolutions to genertate clusterings of n = ns_start to ns_end clusters."/>
                <param name="ns_end" type="integer" value="21" min="1" label="The maximum number of leiden clusters clusters" help="Clusterings are calculated with different leiden resolutions to genertate clusterings of n = ns_start to ns_end clusters."/>
                <repeat name="series_auc_borders" title="Calculates nmi AUCs over given borders">
                    <param name="auc_borders_start" type="integer" value="15" min="1" label="Calculates nmi over n ranges auc_borders_start to auc_borders_end" help="Defined border shouldn't exceed values in nmis."/>
                    <param name="auc_borders_end" type="integer" value="20" min="1" label="Calculates nmi over n ranges auc_borders_start to auc_borders_end" help="Defined border shouldn't exceed values in nmis."/>
                </repeat>
            </when>
            <when value="plot_knn_overlap">
                <expand macro="set_ids"/>
                <repeat name="knn" title="Calculate knn graphs for each k">
                    <param name="k" type="integer" value="10" min="1" label="Includes nearest neighbors for all k"/>
                </repeat>
            </when>
            <when value="plot_marker_corr">
                <expand macro="param_markerset"/>
                <expand macro="set_ids"/>
                <param name="per_celltype" type="boolean" truevalue="True" falsevalue="False" label="Wether to return columns with per cell type max correlations" checked="true"/>
                <param name="per_marker" type="boolean" truevalue="True" falsevalue="False" label="Wether to return columns with per marker max correlations" checked="true"/>
                <conditional name="select_per_celltype_min_mean">
                    <param name="use_per_celltype_min_mean" type="select" label="Add a column for correlation per cell type that only takes into accounts markers with mean expression > per_celltype_min_mean">
                        <option value="False">No</option>
                        <option value="True">Yes</option>
                    </param>
                    <when value="True">
                        <param name="per_celltype_min_mean" type="float" value="0.0" min="0.0" label="Set per_celltype_min_mean"/>
                    </when>
                    <when value="False"/>
                </conditional>
                <conditional name="select_per_marker_min_mean">
                    <param name="use_per_marker_min_mean" type="select" label="Add a column for correlation per cell type that only takes into accounts markers with mean expression > per_celltype_min_mean">
                        <option value="False">No</option>
                        <option value="True">Yes</option>
                    </param>
                    <when value="True">
                        <param name="per_marker_min_mean" type="float" value="0.0" min="0.0" label="Set per_marker_min_mean"/>
                    </when>
                    <when value="False"/>
                </conditional>
            </when>
        </conditional>
        <section name="figure_options" title="Figure Output Options" expanded="false">
            <param argument="dpi" type="integer" value="300" min="1" label="Dpi of figures"/> 
            <param argument="fontsize" type="integer" value="100" min="1" label="Font size of figures"/> 
        </section>
        <expand macro="inputs_common_advanced"/>
    </inputs>
    <outputs>
        <data name="out_png" format="png" from_work_dir="*.png" label="PNG plot from ${tool.name} (${method.method}) on ${on_string}">
            <filter>format == 'png'</filter>
        </data>
        <data name="out_pdf" format="pdf" from_work_dir="*.pdf" label="PDF plot from ${tool.name} (${method.method}) on ${on_string}">
            <filter>format == 'pdf'</filter>
        </data>
        <data name="out_svg" format="svg" from_work_dir="*.svg" label="SVG plot from ${tool.name} (${method.method}) on ${on_string}">
            <filter>format == 'svg'</filter>
        </data>
        <expand macro="hidden_outputs"/>
    </outputs>
    <tests>
        <test expect_num_outputs="2">
            <param name="adata" value="tl.rank_genes_groups.newton-cg.pbmc68k_reduced_240cells.h5ad"/>
            <param name="format" value="png"/>
            <param name="probeset" value="tl.rank_genes_groups.newton-cg.pbmc68k_highly_reduced_marker_filtered_1.tsv"/>
            <param name="header_probeset" value="included"/>
            <param name="nreference" value="30"/>
            <param name="methods_reference" value="random,PCA,DE,HVG"/>
            <param name="genes_key" value="highly_variable"/>
            <param name="obs_key" value="celltype"/>
            <param name="seeds" value="123,124"/>
            <conditional name="method">
                <param name="method" value="plot_summary"/>
                <param name="set_ids" value="all"/>
                <param name="forest_clfs_threshold" value="0.8"/>
                <param name="gene_corr_threshold" value="0.8"/>
                <param name="ns_start" value="5"/>
                <param name="ns_end" value="21"/>
                <repeat name="series_auc_borders">
                    <param name="auc_borders_start" value="7"/>
                    <param name="auc_borders_end" value="14"/>
                </repeat>
                <repeat name="series_auc_borders">
                    <param name="auc_borders_start" value="15"/>
                    <param name="auc_borders_end" value="20"/>
                </repeat>
                <repeat name="knn">
                    <param name="k" value="5"/>
                </repeat>
                <repeat name="knn">
                    <param name="k" value="10"/>
                </repeat>
                <conditional name="select_marker_corr">
                    <param name="use_marker_corr" value="True"/>
                    <param name="markerset" value="marker.tsv"/>
                    <param name="header_markerset" value="not_included"/>
                    <param name="per_celltype" value="True"/>
                    <param name="per_marker" value="True"/>
                    <conditional name="select_per_celltype_min_mean">
                        <param name="use_per_celltype_min_mean" value="False"/>
                    </conditional>
                    <conditional name="select_per_marker_min_mean">
                        <param name="use_per_marker_min_mean" value="True"/>
                        <param name="per_marker_min_mean" value="0.025"/>
                    </conditional>
                </conditional>
            </conditional>
            <param name="show_log" value="true" />
            <output name="hidden_output">
                <assert_contents>
                    <has_text_matching expression="'name': 'probeset',"/>
                    <has_text_matching expression="'per_celltype': True,"/>
                    <has_text_matching expression="'per_marker': True,"/>
                    <has_text_matching expression="'per_marker_min_mean': 0.025"/>
                    <has_text_matching expression="evaluator.plot_summary"/>
                    <has_text_matching expression="save='plot.png',"/>
                </assert_contents>
            </output>
            <output name="out_png">
                <assert_contents>
                    <has_image_width width="3253" delta="2"/>
                    <has_image_height height="1446" delta="2"/>
                </assert_contents>
            </output>
        </test>
        <test expect_num_outputs="2">
            <param name="adata" value="tl.rank_genes_groups.newton-cg.pbmc68k_reduced_240cells.h5ad"/>
            <param name="format" value="png"/>
            <param name="probeset" value="tl.rank_genes_groups.newton-cg.pbmc68k_highly_reduced_marker_filtered_1.tsv"/>
            <param name="header_probeset" value="included"/>
            <param name="nreference" value="30"/>
            <param name="methods_reference" value="random,PCA,DE,HVG"/>
            <param name="genes_key" value="highly_variable"/>
            <param name="obs_key" value="celltype"/>
            <param name="seeds" value="123,124"/>
            <conditional name="method">
                <param name="method" value="plot_confusion_matrix"/>
                <param name="set_ids" value="all"/>
            </conditional>
            <param name="show_log" value="true" />
            <output name="hidden_output">
                <assert_contents>
                    <has_text_matching expression="'name': 'probeset',"/>
                    <has_text_matching expression="'ct_key': 'celltype',"/>
                    <has_text_matching expression="evaluator.plot_confusion_matrix"/>
                    <has_text_matching expression="save='plot.png',"/>
                </assert_contents>
            </output>
            <output name="out_png">
                <assert_contents>
                    <has_image_width width="4560" delta="2"/>
                    <has_image_height height="1859" delta="2"/>
                </assert_contents>
            </output>
        </test>
        <test expect_num_outputs="2">
            <param name="adata" value="tl.rank_genes_groups.newton-cg.pbmc68k_reduced_240cells.h5ad"/>
            <param name="format" value="png"/>
            <param name="probeset" value="tl.rank_genes_groups.newton-cg.pbmc68k_highly_reduced_marker_filtered_1.tsv"/>
            <param name="header_probeset" value="included"/>
            <param name="nreference" value="30"/>
            <param name="methods_reference" value="random,PCA,DE,HVG"/>
            <param name="genes_key" value="highly_variable"/>
            <param name="obs_key" value="celltype"/>
            <param name="seeds" value="123,124"/>
            <conditional name="method">
                <param name="method" value="plot_coexpression"/>
                <param name="set_ids" value="all"/>
            </conditional>
            <param name="show_log" value="true" />
            <output name="hidden_output">
                <assert_contents>
                    <has_text_matching expression="'name': 'probeset',"/>
                    <has_text_matching expression="evaluator.plot_coexpression"/>
                    <has_text_matching expression="save='plot.png',"/>
                </assert_contents>
            </output>
            <output name="out_png">
                <assert_contents>
                    <has_image_width width="5412" delta="2"/>
                    <has_image_height height="3463" delta="2"/>
                </assert_contents>
            </output>
        </test>
        <test expect_num_outputs="2">
            <param name="adata" value="tl.rank_genes_groups.newton-cg.pbmc68k_reduced_240cells.h5ad"/>
            <param name="format" value="png"/>
            <param name="probeset" value="tl.rank_genes_groups.newton-cg.pbmc68k_highly_reduced_marker_filtered_1.tsv"/>
            <param name="header_probeset" value="included"/>
            <param name="nreference" value="30"/>
            <param name="methods_reference" value="random,PCA,DE,HVG"/>
            <param name="genes_key" value="highly_variable"/>
            <param name="obs_key" value="celltype"/>
            <param name="seeds" value="123,124"/>
            <conditional name="method">
                <param name="method" value="plot_cluster_similarity"/>
                <param name="set_ids" value="all"/>
                <param name="ns_start" value="3"/>
                <param name="ns_end" value="20"/>
                <repeat name="series_auc_borders">
                    <param name="auc_borders_start" value="7"/>
                    <param name="auc_borders_end" value="14"/>
                </repeat>
                <repeat name="series_auc_borders">
                    <param name="auc_borders_start" value="15"/>
                    <param name="auc_borders_end" value="20"/>
                </repeat>
            </conditional>
            <param name="show_log" value="true" />
            <output name="hidden_output">
                <assert_contents>
                    <has_text_matching expression="'name': 'probeset',"/>
                    <has_text_matching expression="evaluator.plot_cluster_similarity"/>
                    <has_text_matching expression="save='plot.png',"/>
                </assert_contents>
            </output>
            <output name="out_png">
                <assert_contents>
                    <has_image_width width="3223" delta="2"/>
                    <has_image_height height="1406" delta="2"/>
                </assert_contents>
            </output>
        </test>
        <test expect_num_outputs="2">
            <param name="adata" value="tl.rank_genes_groups.newton-cg.pbmc68k_reduced_240cells.h5ad"/>
            <param name="format" value="png"/>
            <param name="probeset" value="tl.rank_genes_groups.newton-cg.pbmc68k_highly_reduced_marker_filtered_1.tsv"/>
            <param name="header_probeset" value="included"/>
            <param name="nreference" value="30"/>
            <param name="methods_reference" value="random,PCA,DE,HVG"/>
            <param name="genes_key" value="highly_variable"/>
            <param name="obs_key" value="celltype"/>
            <param name="seeds" value="123,124"/>
            <conditional name="method">
                <param name="method" value="plot_knn_overlap"/>
                <param name="set_ids" value="all"/>
                <repeat name="knn">
                    <param name="k" value="6"/>
                </repeat>
                <repeat name="knn">
                    <param name="k" value="11"/>
                </repeat>
            </conditional>
            <param name="show_log" value="true" />
            <output name="hidden_output">
                <assert_contents>
                    <has_text_matching expression="'name': 'probeset',"/>
                    <has_text_matching expression="evaluator.plot_knn_overlap"/>
                    <has_text_matching expression="save='plot.png',"/>
                </assert_contents>
            </output>
            <output name="out_png">
                <assert_contents>
                    <has_image_width width="3223" delta="2"/>
                    <has_image_height height="1406" delta="2"/>
                </assert_contents>
            </output>
        </test>
        <test expect_num_outputs="2">
            <param name="adata" value="tl.rank_genes_groups.newton-cg.pbmc68k_reduced_240cells.h5ad"/>
            <param name="format" value="png"/>
            <param name="probeset" value="tl.rank_genes_groups.newton-cg.pbmc68k_highly_reduced_marker_filtered_1.tsv"/>
            <param name="header_probeset" value="included"/>
            <param name="nreference" value="50"/>
            <param name="methods_reference" value="random,PCA,DE,HVG"/>
            <param name="genes_key" value="highly_variable"/>
            <param name="obs_key" value="celltype"/>
            <conditional name="method">
                <param name="method" value="plot_marker_corr"/>
                <param name="set_ids" value="all"/>
                <param name="use_marker_corr" value="True"/>
                <param name="markerset" value="marker.tsv"/>
                <param name="header_markerset" value="not_included"/>
                <param name="per_celltype" value="True"/>
                <param name="per_marker" value="True"/>
                <conditional name="select_per_celltype_min_mean">
                    <param name="use_per_celltype_min_mean" value="False"/>
                </conditional>
                <conditional name="select_per_marker_min_mean">
                    <param name="use_per_marker_min_mean" value="True"/>
                    <param name="per_marker_min_mean" value="0.025"/>
                </conditional>
            </conditional>
            <param name="show_log" value="true" />
            <output name="hidden_output">
                <assert_contents>
                    <has_text_matching expression="'name': 'probeset',"/>
                    <has_text_matching expression="'per_celltype': True,"/>
                    <has_text_matching expression="'per_marker': True,"/>
                    <has_text_matching expression="'per_marker_min_mean': 0.025"/>
                    <has_text_matching expression="evaluator.plot_marker_corr"/>
                    <has_text_matching expression="save='plot.png'"/>
                </assert_contents>
            </output>
            <output name="out_png">
                <assert_contents>
                    <has_image_width width="5064" delta="2"/>
                    <has_image_height height="4554" delta="2"/>
                </assert_contents>
            </output>
        </test>
    </tests>
    <help><![CDATA[
Probe set evaluation for single-cell sequencing data using spapros.
============================================================================================================

Spapros is a python package that provides a pipeline for probe set selection and evaluation for targeted spatial transcriptomics data.

Key Features:
* Select probe sets for spatial transcriptomics which identify cell types of interest, capture general transcriptomic variation, and incorporate prior knowledge
* Evaluate probe sets with an extensive pipeline

Further documentation can be found here: https://spapros.readthedocs.io/en/latest/index.html.

    ]]></help>
    <expand macro="citations"/>
</tool>