comparison filter.xml @ 6:a97abb8cd15b draft

"planemo upload for repository https://github.com/galaxyproject/tools-iuc/tree/master/tools/scanpy/ commit 5a90fd345b43ca12366f4475f4cfd88ef197e452"

5:8b9610ab366a

                <param argument="total_counts" type="integer" min="0" optional="true" label="Target total counts" help="If the count matrix has more than total_counts it will be downsampled to have this number."/>
                <param argument="random_state" type="integer" value="0" label="Random seed to change subsampling"/>
                <param argument="replace" type="boolean" truevalue="True" falsevalue="False" checked="false" label="Sample the counts with replacement?"/>
            </when>
        </conditional>
    </inputs>
    <outputs>
        <expand macro="anndata_outputs"/>
    </outputs>
    <tests>
        <test>
            <!-- test 1 -->
            <param name="adata" value="krumsiek11.h5ad" />
            <conditional name="method">
                <param name="method" value="pp.filter_cells"/>
                <conditional name="filter">
                    <param name="filter" value="min_counts"/>
                    <param name="min_counts" value="3"/>
                </conditional>
            </conditional>
            <assert_stdout>
                <has_text_matching expression="sc.pp.filter_cells"/>
                <has_text_matching expression="min_counts=3"/>
            </assert_stdout>
            <output name="anndata_out" file="pp.filter_cells.krumsiek11-min_counts.h5ad" ftype="h5ad" compare="sim_size"/>
        </test>
        <test>
            <!-- test 2 -->
            <param name="adata" value="krumsiek11.h5ad" />
            <conditional name="method">
                <param name="method" value="pp.filter_cells"/>
                <conditional name="filter">
                    <param name="filter" value="max_genes"/>
                    <param name="max_genes" value="100"/>
                </conditional>
            </conditional>
            <assert_stdout>
                <has_text_matching expression="sc.pp.filter_cells"/>
                <has_text_matching expression="adata"/>
                <has_text_matching expression="max_genes=100"/>
            </assert_stdout>
            <output name="anndata_out" file="pp.filter_cells.krumsiek11-max_genes.h5ad" ftype="h5ad" compare="sim_size"/>
        </test>
        <test>
            <!-- test 3 -->
            <param name="adata" value="krumsiek11.h5ad" />
            <conditional name="method">
                <param name="method" value="pp.filter_genes"/>
                <conditional name="filter">
                    <param name="filter" value="min_counts"/>
                    <param name="min_counts" value="3"/>
                </conditional>
            </conditional>
            <assert_stdout>
                <has_text_matching expression="sc.pp.filter_genes"/>
                <has_text_matching expression="min_counts=3"/>
            </assert_stdout>
            <output name="anndata_out" file="pp.filter_genes.krumsiek11-min_counts.h5ad" ftype="h5ad" compare="sim_size"/>
        </test>
        <test>
            <!-- test 4 -->
            <param name="adata" value="tl.rank_genes_groups.krumsiek11.h5ad" />
            <conditional name="method">
                <param name="method" value="tl.filter_rank_genes_groups"/>
                <param name="key" value="rank_genes_groups"/>
                <param name="use_raw" value="False"/>
                <param name="log" value="False"/>
                <param name="key_added" value="rank_genes_groups_filtered"/>
                <param name="min_in_group_fraction" value="0.25"/>
                <param name="max_out_group_fraction" value="0.5"/>
                <param name="min_fold_change" value="3"/>
            </conditional>
            <assert_stdout>
                <has_text_matching expression="tl.filter_rank_genes_groups"/>
                <has_text_matching expression="key='rank_genes_groups'"/>
                <has_text_matching expression="use_raw=False"/>
                <has_text_matching expression="log=False"/>
                <has_text_matching expression="key_added='rank_genes_groups_filtered'"/>
                <has_text_matching expression="min_in_group_fraction=0.25"/>
                <has_text_matching expression="max_out_group_fraction=0.5"/>
                <has_text_matching expression="min_fold_change=3"/>
            </assert_stdout>
            <output name="anndata_out" file="pp.filter_rank_genes_groups.h5ad" ftype="h5ad" compare="sim_size"/>
        </test>
        <test>
            <!-- test 5 -->
            <param name="adata" value="blobs.h5ad"/>
            <conditional name="method">
                <param name="method" value="pp.highly_variable_genes"/>
                <conditional name="flavor">
                    <param name="flavor" value="seurat"/>

                    <param name="min_disp" value="0.5"/>
                </conditional>
                <param name="n_bins" value="20"/>
                <param name="subset" value="false"/>
            </conditional>
            <assert_stdout>
                <has_text_matching expression="sc.pp.highly_variable_genes"/>
                <has_text_matching expression="flavor='seurat'"/>
                <has_text_matching expression="min_mean=0.0125"/>
                <has_text_matching expression="max_mean=3"/>
                <has_text_matching expression="min_disp=0.5"/>
                <has_text_matching expression="n_bins=20"/>
                <has_text_matching expression="subset=False"/>
            </assert_stdout>
            <output name="anndata_out" file="pp.highly_variable_genes.seurat.blobs.h5ad" ftype="h5ad" compare="sim_size"/>
        </test>
        <test>
            <!-- test 6 -->
            <param name="adata" value="krumsiek11.h5ad" />
            <conditional name="method">
                <param name="method" value="pp.highly_variable_genes"/>
                <conditional name="flavor">
                    <param name="flavor" value="cell_ranger"/>
                    <param name="n_top_genes" value="2"/>
                </conditional>
                <param name="n_bins" value="20"/>
                <param name="subset" value="true"/>
            </conditional>
            <assert_stdout>
                <has_text_matching expression="sc.pp.highly_variable_genes"/>
                <has_text_matching expression="flavor='cell_ranger'"/>
                <has_text_matching expression="n_top_genes=2"/>
                <has_text_matching expression="n_bins=20"/>
                <has_text_matching expression="subset=True"/>
            </assert_stdout>
            <output name="anndata_out" file="pp.highly_variable_genes.krumsiek11-cell_ranger.h5ad" ftype="h5ad" compare="sim_size"/>
        </test>
        <test>
            <!-- test 7 -->
            <param name="adata" value="krumsiek11.h5ad" />
            <conditional name="method">
                <param name="method" value="pp.subsample"/>
                <conditional name="type">
                    <param name="type" value="fraction" />
                    <param name="fraction" value="0.5"/>
                </conditional>
                <param name="random_state" value="0"/>
            </conditional>
            <assert_stdout>
                <has_text_matching expression="sc.pp.subsample"/>
                <has_text_matching expression="fraction=0.5"/>
                <has_text_matching expression="random_state=0"/>
            </assert_stdout>
            <output name="anndata_out" file="pp.subsample.krumsiek11_fraction.h5ad" ftype="h5ad" compare="sim_size"/>
        </test>
        <test>
            <!-- test 8 -->
            <param name="adata" value="krumsiek11.h5ad" />
            <conditional name="method">
                <param name="method" value="pp.subsample"/>
                <conditional name="type">
                    <param name="type" value="n_obs" />
                    <param name="n_obs" value="10"/>
                </conditional>
                <param name="random_state" value="0"/>
            </conditional>
            <assert_stdout>
                <has_text_matching expression="sc.pp.subsample"/>
                <has_text_matching expression="n_obs=10"/>
                <has_text_matching expression="random_state=0"/>
            </assert_stdout>
            <output name="anndata_out" file="pp.subsample.krumsiek11_n_obs.h5ad" ftype="h5ad" compare="sim_size"/>
        </test>
        <test>
            <!-- test 9 -->
            <param name="adata" value="random-randint.h5ad" />
            <conditional name="method">
                <param name="method" value="pp.downsample_counts"/>
                <param name="total_counts" value="20000"/>
                <param name="random_state" value="0"/>
                <param name="replace" value="false"/>
            </conditional>
            <assert_stdout>
                <has_text_matching expression="sc.pp.downsample_counts"/>
                <has_text_matching expression="total_counts=20000"/>
                <has_text_matching expression="random_state=0"/>
                <has_text_matching expression="replace=False"/>
            </assert_stdout>
            <output name="anndata_out" file="pp.downsample_counts.random-randint.h5ad" ftype="h5ad" compare="sim_size"/>
        </test>
    </tests>
    <help><![CDATA[

6:a97abb8cd15b

                <param argument="total_counts" type="integer" min="0" optional="true" label="Target total counts" help="If the count matrix has more than total_counts it will be downsampled to have this number."/>
                <param argument="random_state" type="integer" value="0" label="Random seed to change subsampling"/>
                <param argument="replace" type="boolean" truevalue="True" falsevalue="False" checked="false" label="Sample the counts with replacement?"/>
            </when>
        </conditional>
        <expand macro="inputs_common_advanced"/>
    </inputs>
    <outputs>
        <expand macro="anndata_outputs"/>
    </outputs>
    <tests>
        <test>
            <!-- test 0 -->
            <param name="adata" value="krumsiek11.h5ad" />
            <conditional name="method">
                <param name="method" value="pp.filter_cells"/>
                <conditional name="filter">
                    <param name="filter" value="min_counts"/>
                    <param name="min_counts" value="3"/>
                </conditional>
            </conditional>
            <assert_stdout>
                <has_text_matching expression="336 × 11"/>
            </assert_stdout>
            <section name="advanced_common">
                <param name="show_log" value="true" />
            </section>
            <output name="hidden_output">
                <assert_contents>
                    <has_text_matching expression="sc.pp.filter_cells"/>
                    <has_text_matching expression="min_counts=3"/>
                </assert_contents>
            </output>
            <output name="anndata_out" file="pp.filter_cells.krumsiek11-min_counts.h5ad" ftype="h5ad" compare="sim_size"/>
        </test>
        <test>
            <!-- test 1 -->
            <param name="adata" value="krumsiek11.h5ad" />
            <conditional name="method">
                <param name="method" value="pp.filter_cells"/>
                <conditional name="filter">
                    <param name="filter" value="max_genes"/>
                    <param name="max_genes" value="100"/>
                </conditional>
            </conditional>
            <section name="advanced_common">
                <param name="show_log" value="true" />
            </section>
            <output name="hidden_output">
                <assert_contents>
                    <has_text_matching expression="sc.pp.filter_cells"/>
                    <has_text_matching expression="adata"/>
                    <has_text_matching expression="max_genes=100"/>
                </assert_contents>
            </output>
            <output name="anndata_out" file="pp.filter_cells.krumsiek11-max_genes.h5ad" ftype="h5ad" compare="sim_size"/>
        </test>
        <test>
            <!-- test 2 -->
            <param name="adata" value="krumsiek11.h5ad" />
            <conditional name="method">
                <param name="method" value="pp.filter_genes"/>
                <conditional name="filter">
                    <param name="filter" value="min_counts"/>
                    <param name="min_counts" value="3"/>
                </conditional>
            </conditional>
            <section name="advanced_common">
                <param name="show_log" value="true" />
            </section>
            <output name="hidden_output">
                <assert_contents>
                    <has_text_matching expression="sc.pp.filter_genes"/>
                    <has_text_matching expression="min_counts=3"/>
                </assert_contents>
            </output>
            <output name="anndata_out" file="pp.filter_genes.krumsiek11-min_counts.h5ad" ftype="h5ad" compare="sim_size"/>
        </test>
        <!-- <test> -->
        <!--     <!-\- test 3 -\-> -->
        <!--     <!-\- Input dataset appears to be missing rank_genes_groups key... -\-> -->
        <!--     <param name="adata" value="tl.rank_genes_groups.krumsiek11.h5ad" /> -->
        <!--     <conditional name="method"> -->
        <!--         <param name="method" value="tl.filter_rank_genes_groups"/> -->
        <!--         <param name="key" value="rank_genes_groups"/> -->
        <!--         <param name="use_raw" value="False"/> -->
        <!--         <param name="log" value="False"/> -->
        <!--         <param name="key_added" value="rank_genes_groups_filtered"/> -->
        <!--         <param name="min_in_group_fraction" value="0.25"/> -->
        <!--         <param name="max_out_group_fraction" value="0.5"/> -->
        <!--         <param name="min_fold_change" value="3"/> -->
        <!--     </conditional> -->
        <!--     <output name="hidden_output"> -->
        <!--         <assert_contents> -->
        <!--             <has_text_matching expression="tl.filter_rank_genes_groups"/> -->
        <!--             <has_text_matching expression="key='rank_genes_groups'"/> -->
        <!--             <has_text_matching expression="use_raw=False"/> -->
        <!--             <has_text_matching expression="log=False"/> -->
        <!--             <has_text_matching expression="key_added='rank_genes_groups_filtered'"/> -->
        <!--             <has_text_matching expression="min_in_group_fraction=0.25"/> -->
        <!--             <has_text_matching expression="max_out_group_fraction=0.5"/> -->
        <!--             <has_text_matching expression="min_fold_change=3"/> -->
        <!--         </assert_contents> -->
        <!--     </output> -->
        <!--     <output name="anndata_out" file="pp.filter_rank_genes_groups.h5ad" ftype="h5ad" compare="sim_size"/> -->
        <!-- </test> -->
        <test>
            <!-- test 4 -->
            <param name="adata" value="blobs.h5ad"/>
            <conditional name="method">
                <param name="method" value="pp.highly_variable_genes"/>
                <conditional name="flavor">
                    <param name="flavor" value="seurat"/>

                    <param name="min_disp" value="0.5"/>
                </conditional>
                <param name="n_bins" value="20"/>
                <param name="subset" value="false"/>
            </conditional>
            <section name="advanced_common">
                <param name="show_log" value="true" />
            </section>
            <output name="hidden_output">
                <assert_contents>
                    <has_text_matching expression="sc.pp.highly_variable_genes"/>
                    <has_text_matching expression="flavor='seurat'"/>
                    <has_text_matching expression="min_mean=0.0125"/>
                    <has_text_matching expression="max_mean=3"/>
                    <has_text_matching expression="min_disp=0.5"/>
                    <has_text_matching expression="n_bins=20"/>
                    <has_text_matching expression="subset=False"/>
                </assert_contents>
            </output>
            <output name="anndata_out" file="pp.highly_variable_genes.seurat.blobs.h5ad" ftype="h5ad" compare="sim_size"/>
        </test>
        <test>
            <!-- test 5 -->
            <param name="adata" value="krumsiek11.h5ad" />
            <conditional name="method">
                <param name="method" value="pp.highly_variable_genes"/>
                <conditional name="flavor">
                    <param name="flavor" value="cell_ranger"/>
                    <param name="n_top_genes" value="2"/>
                </conditional>
                <param name="n_bins" value="20"/>
                <param name="subset" value="true"/>
            </conditional>
            <section name="advanced_common">
                <param name="show_log" value="true" />
            </section>
            <output name="hidden_output">
                <assert_contents>
                    <has_text_matching expression="sc.pp.highly_variable_genes"/>
                    <has_text_matching expression="flavor='cell_ranger'"/>
                    <has_text_matching expression="n_top_genes=2"/>
                    <has_text_matching expression="n_bins=20"/>
                    <has_text_matching expression="subset=True"/>
                </assert_contents>
            </output>
            <output name="anndata_out" file="pp.highly_variable_genes.krumsiek11-cell_ranger.h5ad" ftype="h5ad" compare="sim_size"/>
        </test>
        <test>
            <!-- test 6 -->
            <param name="adata" value="krumsiek11.h5ad" />
            <conditional name="method">
                <param name="method" value="pp.subsample"/>
                <conditional name="type">
                    <param name="type" value="fraction" />
                    <param name="fraction" value="0.5"/>
                </conditional>
                <param name="random_state" value="0"/>
            </conditional>
            <section name="advanced_common">
                <param name="show_log" value="true" />
            </section>
            <output name="hidden_output">
                <assert_contents>
                    <has_text_matching expression="sc.pp.subsample"/>
                    <has_text_matching expression="fraction=0.5"/>
                    <has_text_matching expression="random_state=0"/>
                </assert_contents>
            </output>
            <output name="anndata_out" file="pp.subsample.krumsiek11_fraction.h5ad" ftype="h5ad" compare="sim_size"/>
        </test>
        <test>
            <!-- test 7 -->
            <param name="adata" value="krumsiek11.h5ad" />
            <conditional name="method">
                <param name="method" value="pp.subsample"/>
                <conditional name="type">
                    <param name="type" value="n_obs" />
                    <param name="n_obs" value="10"/>
                </conditional>
                <param name="random_state" value="0"/>
            </conditional>
            <section name="advanced_common">
                <param name="show_log" value="true" />
            </section>
            <output name="hidden_output">
                <assert_contents>
                    <has_text_matching expression="sc.pp.subsample"/>
                    <has_text_matching expression="n_obs=10"/>
                    <has_text_matching expression="random_state=0"/>
                </assert_contents>
            </output>
            <output name="anndata_out" file="pp.subsample.krumsiek11_n_obs.h5ad" ftype="h5ad" compare="sim_size"/>
        </test>
        <test>
            <!-- test 8 -->
            <param name="adata" value="random-randint.h5ad" />
            <conditional name="method">
                <param name="method" value="pp.downsample_counts"/>
                <param name="total_counts" value="20000"/>
                <param name="random_state" value="0"/>
                <param name="replace" value="false"/>
            </conditional>
            <section name="advanced_common">
                <param name="show_log" value="true" />
            </section>
            <output name="hidden_output">
                <assert_contents>
                    <has_text_matching expression="sc.pp.downsample_counts"/>
                    <has_text_matching expression="total_counts=20000"/>
                    <has_text_matching expression="random_state=0"/>
                    <has_text_matching expression="replace=False"/>
                </assert_contents>
            </output>
            <output name="anndata_out" file="pp.downsample_counts.random-randint.h5ad" ftype="h5ad" compare="sim_size"/>
        </test>
    </tests>
    <help><![CDATA[