comparison filter_cells.xml @ 0:e63bd8f13679 draft

planemo upload for repository https://github.com/ARTbio/tools-artbio/tree/master/tools/gsc_filter_cells commit 09dcd74dbc01f448518cf3db3e646afb0675a6fe

0:e63bd8f13679

<tool id="filter_cells" name="Filter cells data" version="0.9.0">
    <description>on total aligned reads and/or number of detected genes</description>
    <requirements>
        <requirement type="package" version="1.3.2=r3.3.2_0">r-optparse</requirement>
        <requirement type="package" version="2.2.1=r3.3.2_0">r-ggplot2</requirement>
    </requirements>
    <stdio>
        <exit_code range="1:" level="fatal" description="Tool exception" />
    </stdio>
    <command detect_errors="exit_code"><![CDATA[ 
        Rscript $__tool_directory__/filter_cells.R 
            -f $input 
            --sep 
            #if $sep == 'tab':
              'tab'
            #elif $sep == 'comma':
              'comma'
            #end if
            --percentile_genes '$percentile_genes'
            --percentile_counts '$percentile_counts'
            --absolute_genes '$absolute_genes'
            --absolute_counts '$absolute_counts'
            --manage_cutoffs
            #if $manage_cutoffs == 'union':
              'union'
            #elif $manage_cutoffs == 'intersect':
              'intersect'
            #end if
            --pdfplot $pdfplot
            --output $output
            --output_metada $output_metada
]]></command>
    <inputs>
        <param name="input" type="data" format="txt,tabular" label="Raw counts of expression data"/>
        <param name="sep" type="select" label="Indicate column separator">
            <option value="tab" selected="true">Tabs</option>
            <option value="comma">Comma</option>
        </param>
        <param name="percentile_genes" value="0" type="integer" label="Detected Genes Percentile Threshold [integer] %"
               help="Cells with % of detected genes below this threshold will be filtered out. Leave at 0 for no filtering" />
        <param name="percentile_counts" value="0" type="integer" label="Aligned read count Percentile Threshold [integer] %"
               help="Cells with number of aligned reads below this % threshold will be filtered out. Leave at 0 for no filtering" />
        <param name="absolute_genes" value="0" type="integer" label="Absolute number of Detected Genes Threshold [integer]"
               help="Cells with number of detected genes below this threshold will be filtered out. Leave at 0 for no filtering" />
        <param name="absolute_counts" value="0" type="integer" label="Absolute number of aligned read Threshold [integer]"
               help="Cells with number of aligned reads below this absolute threshold will be filtered out. Leave at 0 for no filtering" />
        <param name="manage_cutoffs" type="select" label=" filter out intersection or union of cutoffs"
               help="If you use two cutoffs on number of detected genes and number of aligned reads, respectively, there are two options
               for using these cutoffs in filtering: either excluding items that are below one or the other threshold (union) or
               excluding items that are below both thresholds (intersection)" >
            <option value="union" selected="true">Union of cutoffs</option>
            <option value="intersect">Intersection of cutoffs</option>
        </param>
    </inputs>
    <outputs>
        <data name="pdfplot" format="pdf" label="Plots from ${on_string}" />
        <data name="output" format="tabular" label="Cell data filtered with ${on_string}" />
        <data name="output_metada" format="tabular" label="Cell metadata filtered with ${on_string}" />
    </outputs>
    <tests>
        <test>
            <param name="input" value="input.csv" ftype="txt"/>
            <param name="sep" value='comma' />
            <param name="percentile_genes" value="20"/>
            <param name="percentile_counts" value="20"/>
            <output name="pdfplot" file="percentile_gene-and-counts.pdf" ftype="pdf"/>
            <output name="output" file="percentile_gene-and-counts.tab" ftype="tabular"/>
            <output name="output_metada" file="percentile_gene-and-counts.meta" ftype="tabular"/>
        </test>
        <test>
            <param name="input" value="input.csv" ftype="txt"/>
            <param name="sep" value='comma' />
            <param name="percentile_genes" value="20"/>
            <param name="percentile_counts" value="20"/>
            <param name="manage_cutoffs" value="intersect"/>
            <output name="pdfplot" file="intersect_percentile_gene-and-counts.pdf" ftype="pdf"/>
            <output name="output" file="intersect_percentile_gene-and-counts.tab" ftype="tabular"/>
            <output name="output_metada" file="intersect_percentile_gene-and-counts.meta" ftype="tabular"/>
        </test>
       <test>
            <param name="input" value="input.tsv" ftype="txt"/>
            <param name="sep" value='tab' />
            <param name="percentile_genes" value="20"/>
            <param name="percentile_counts" value="20"/>
            <output name="pdfplot" file="percentile_gene-and-counts.pdf" ftype="pdf"/>
            <output name="output" file="percentile_gene-and-counts.tab" ftype="tabular"/>
            <output name="output_metada" file="percentile_gene-and-counts.meta" ftype="tabular"/>
        </test>
        <test>
            <param name="input" value="input.csv" ftype="txt"/>
            <param name="sep" value='comma' />
            <param name="percentile_genes" value="20"/>
            <output name="pdfplot" file="percentile_gene-only.pdf" ftype="pdf"/>
            <output name="output" file="percentile_gene-only.tab" ftype="tabular"/>
            <output name="output_metada" file="percentile_gene-only.meta" ftype="tabular"/>
        </test>
        <test>
            <param name="input" value="input.csv" ftype="txt"/>
            <param name="sep" value='comma' />
            <param name="percentile_counts" value="20"/>
            <output name="pdfplot" file="percentile_counts-only.pdf" ftype="pdf"/>
            <output name="output" file="percentile_counts-only.tab" ftype="tabular"/>
            <output name="output_metada" file="percentile_counts-only.meta" ftype="tabular"/>
        </test>
        <test>
            <param name="input" value="input.csv" ftype="txt"/>
            <param name="sep" value='comma' />
            <output name="pdfplot" file="no-filtering.pdf" ftype="pdf"/>
            <output name="output" file="no-filtering.tab" ftype="tabular"/>
            <output name="output_metada" file="no-filtering.meta" ftype="tabular"/>
        </test>
        <test>
            <param name="input" value="input.csv" ftype="txt"/>
            <param name="sep" value='comma' />
            <param name="absolute_genes" value="5"/>
            <param name="absolute_counts" value="1000"/>
            <output name="pdfplot" file="absolute_gene-and-counts.pdf" ftype="pdf"/>
            <output name="output" file="absolute_gene-and-counts.tab" ftype="tabular"/>
            <output name="output_metada" file="absolute_gene-and-counts.meta" ftype="tabular"/>
        </test>
        <test>
            <param name="input" value="input.csv" ftype="txt"/>
            <param name="sep" value='comma' />
            <param name="absolute_genes" value="5"/>
            <output name="pdfplot" file="absolute_gene-only.pdf" ftype="pdf"/>
            <output name="output" file="absolute_gene-only.tab" ftype="tabular"/>
            <output name="output_metada" file="absolute_gene-only.meta" ftype="tabular"/>
        </test>
        <test>
            <param name="input" value="input.csv" ftype="txt"/>
            <param name="sep" value='comma' />
            <param name="absolute_counts" value="1000"/>
            <output name="pdfplot" file="absolute_counts-only.pdf" ftype="pdf"/>
            <output name="output" file="absolute_counts-only.tab" ftype="tabular"/>
            <output name="output_metada" file="absolute_counts-only.meta" ftype="tabular"/>
        </test>
    </tests>
    <help>

**What it does**

The tools takes a table of gene (rows) expression values (as number of reads aligned to genes)
in single cell RNAseq sequencing libraries (columns) and filters out cells with low number
of detected genes and/or cells with low number of aligned reads.

Cutoffs can be applied to absolute numbers of aligned reads or of detected genes, or to
percentile thresholds for these variables.

For both absolute or percentile thresholds, only cells exclusively below
these threshold are excluded (cell cutoffs do not include the threshold values).

If you choose to combine cutoffs for both the number of detected genes
and the total number of aligned reads, then you have 2 options: either exclude libraries that
do not satisfy one OR the other threshold (Union) or exclude libraries that do not satisfy
both thresholds (Intersection).

Specifying a value both for an absolute and an percentile threshold of a variable
(Number of detected genes or Number of aligned counts) is not consistent. In this
situation, the tools *does not* filter cells with respect to the corresponding variable threshold.
If a 0 is applied both for an absolute and an percentile threshold of a variable, then
this variable is not used to filter the cells.

The tools returns a gene expression table for cells that were retained, a metadata table
that contains numbers of detected genes and aligned reads for retained cell library and
a pdf file with three plots illustrating the performed filtering operation.

**Input**

A table of comma (csv) or tabulation (tsv) separated expression values, in number (integers)
of reads aligned to genes.
Gene names should be in the first column and cell names should be in the first row.
Note that in a number of a csv files, header of the gene column is omitted, resulting in
a first row with one item less than in other rows. This is handled by the tool that
recognises this situation.

    </help>
    <citations>
        <citation type="bibtex">
        @Manual{,
             title = {R: A Language and Environment for Statistical Computing},
             author = {{R Core Team}},
             organization = {R Foundation for Statistical Computing},
             address = {Vienna, Austria},
             year = {2014},
             url = {http://www.R-project.org/},
        }
        </citation>
    </citations>
</tool>