<tool id="scmap_index_cell" name="scmap index cells" version="@TOOL_VERSION@+galaxy1" profile="@PROFILE@">
    <description>creates a cell index for a dataset to enable fast approximate nearest neighbour search</description>
    <macros>
        <import>scmap_macros.xml</import>
    </macros>
    <expand macro="requirements" />
    <command detect_errors="exit_code"><![CDATA[
        scmap-preprocess-sce.R --input-object "${input_single_cell_experiment}" --output-sce-object sce_object_preprocessed.rds &&
        scmap-index-cell.R --input-object-file sce_object_preprocessed.rds --number-chunks '$n_chunks' --number-clusters '$n_clusters' --output-object-file '$output_single_cell_experiment' --random-seed '$random_seed'
        #if $train_id
        --train-id '${train_id}' 
        #end if
    ]]></command>
    <inputs>
        <param type="data" name="input_single_cell_experiment" label="SingleCellExperiment object" format="rdata" help="File with serialized SingleCellExperiment object as produced by 'scmap select features'" />
        <param name="n_chunks" type="integer" label="Number of chunks" value='' help="Number of chunks into which the expr matrix is split" />
        <param name="n_clusters" type="integer" label="Number of clusters" value='' help="Number of clusters per group for k-means clustering" />
        <param type="text" name="train_id" label="Dataset ID" help="ID of the training dataset" />
        <param name="random_seed" type="integer" label="Random seed" value='1' help="scmap-cell contains k-means step which makes it stochastic, i.e. running it multiple times will provide slightly different results. A fixed random seed ensures reproducibility." />
    </inputs>
    <outputs>
        <data name="output_single_cell_experiment" format="rdata" />
    </outputs>
    <tests>
        <test>
            <param name="n_chunks" value="50" />
            <param name="n_clusters" value="9" />
            <param name="input_single_cell_experiment" value="select_features.rds" ftype="rdata"/>
            <output name="output_single_cell_experiment" file="index_cell.rds" compare="sim_size"/>
        </test>
    </tests>
    <help><![CDATA[
    @HELP@
    
    @VERSION_HISTORY@
    ]]></help>
    <expand macro="citations" />
</tool>