<tool id="hicstuff_pipeline" name="hicstuff full pipeline" version="@TOOL_VERSION@+galaxy@VERSION_SUFFIX@" profile="@PROFILE@">
    <description>generates a Hi-C contact matrix</description>
    <macros>
        <import>macros.xml</import>
        <token name="@VERSION_SUFFIX@">1</token>
    </macros>
    <expand macro="requirements" />
    <command detect_errors="exit_code"><![CDATA[
hicstuff pipeline 
    --genome '$genome' 
    --outdir results
    --aligner $aligner
    $circular
    $duplicates
    --enzyme '$enzyme'
    $filter
    --mapping $mapping
    --matfmt $matfmt
    --quality-min $quality_min
    --size $size
    --threads "\${GALAXY_SLOTS:-1}"
    #if $paired_cond.paired_select == "paired"
        '$paired_cond.reads.forward'
        '$paired_cond.reads.reverse'
    #else
        '$forward_reads'
        '$reverse_reads'
    #end if
    ]]></command>
    <inputs>
        <param type="data" name="genome" format="fasta,fasta.gz" label="Genome fasta file"/>
        <conditional name="paired_cond">
            <param name="paired_select" type="select" label="Paired reads">
                <option value="paired">In a dataset pair</option>
                <option value="separate">In two separate datasets</option>
            </param>
            <when value="paired">
                <param name="reads" type="data_collection" collection_type="paired" format="fastqsanger,fastqsanger.gz" label="Paired reads"/>
            </when>
            <when value="separate">
                <param name="forward_reads" type="data" format="fastqsanger,fastqsanger.gz" label="Forward reads"/>
                <param name="reverse_reads" type="data" format="fastqsanger,fastqsanger.gz" label="Reverse reads"/>
            </when>
        </conditional>
        <param argument="--aligner" type="select" label="Alignment software to use" help="Minimap2 should only be used for reads > 100 bp">
            <option value="bowtie2" selected="true">bowtie2</option>
            <option value="minimap2">minimap2</option>
            <option value="bwa">bwa</option>
        </param>
        <param argument="--circular" type="boolean" truevalue="--circular" falsevalue="" label="Circular genome"/>
        <param argument="--duplicates" type="boolean" truevalue="--duplicates" falsevalue="" label="Removes PCR duplicates" help="PCR duplicates are defined as sets of pairs having identical mapping positions for both reads."/>
        <param argument="--enzyme" type="text" value="5000" label="Bin size or enzyme" help="Restriction enzyme or 'mnase' if a string, or chunk size (i.e. resolution) if a number. Can also be multiple comma-separated enzymes."/>
        <param argument="--filter" type="boolean" truevalue="--filter" falsevalue="" label="Filters out spurious 3C events, such as self religations or undigested fragments" help="This is only really useful at very fine resolutions (1-2kb) and not needed most of the time. This option is only meaningful when --enzyme is given a restriction enzyme and not a bin size."/>
        <param argument="--mapping" type="select" label="Parameter of mapping" help="'normal': Directly map reads without any process. 'iterative': Map reads iteratively using iteralign, by truncating reads to 20bp and then repeatedly extending to align them. 'cutsite': Cut reads at the religation sites of the given enzyme using cutsite, create new pairs of reads and then align them ; enzyme is required">
            <option value="normal" selected="true">normal</option>
            <option value="iterative">iterative</option>
            <option value="cutsite">cutsite</option>
        </param>
        <param argument="--matfmt" type="select" label="Format of the output sparse matrix" help="Available formats are bg2 (bedgraph2d), graal (graal-compatible plain text COO format) and cool, a binary format that is probably the most appropriate for large genomes.">
            <option value="bg2">bg2</option>
            <option value="cool">cool</option>
            <option value="graal" selected="true">graal</option>
        </param>
        <param argument="--quality-min" type="integer" value="30" label="Minimum mapping quality for selecting contacts"/>
        <param argument="--size" type="integer" value="0" label="Minimum size threshold to consider contigs. Keep all contigs by default."/>
    </inputs>
    <outputs>
        <data name="abs_fragments_contacts_weighted_graal" from_work_dir="./results/abs_fragments_contacts_weighted.txt" format="tabular" label="${tool.name} on ${on_string}: Contact matrix (graal)">
            <filter>matfmt == "graal"</filter>
        </data>
        <data name="abs_fragments_contacts_weighted_bg2" from_work_dir="./results/abs_fragments_contacts_weighted.bg2" format="tabular" label="${tool.name} on ${on_string}: Contact matrix (bg2)">
            <filter>matfmt == "bg2"</filter>
        </data>
        <data name="abs_fragments_contacts_weighted_cool" from_work_dir="./results/abs_fragments_contacts_weighted.cool" format="cool" label="${tool.name} on ${on_string}: Contact matrix (cool)">
            <filter>matfmt == "cool"</filter>
        </data>
        <data name="fragments_list" from_work_dir="./results/fragments_list.txt" format="tabular" label="${tool.name} on ${on_string}: Fragment list"/>
        <data name="info_contigs" from_work_dir="./results/info_contigs.txt" format="tabular" label="${tool.name} on ${on_string}: Contig info"/>
    </outputs>
    <tests>
        <test expect_num_outputs="3">
            <param name="genome" value="seq.fa.gz" />
            <param name="paired_cond|paired_select" value="separate"/>
            <param name="paired_cond|forward_reads" value="sample.reads_for.fastq.gz" />
            <param name="paired_cond|reverse_reads" value="sample.reads_rev.fastq.gz" />
            <output name="abs_fragments_contacts_weighted_graal">
                <assert_contents>
                    <has_n_lines n="77"/>
                    <has_n_columns n="3"/>
                </assert_contents>
            </output>
            <output name="fragments_list">
                <assert_contents>
                    <has_n_lines n="17"/>
                    <has_n_columns n="6"/>
                </assert_contents>
            </output>
            <output name="info_contigs" file="info_contigs.txt"/>
            <assert_stderr>
                <has_text text="Contact map generated" />
            </assert_stderr>
        </test>
        <!-- bg2 format -->
        <test expect_num_outputs="3">
            <param name="genome" value="seq.fa.gz" />
            <param name="paired_cond|paired_select" value="separate"/>
            <param name="paired_cond|forward_reads" value="sample.reads_for.fastq.gz" />
            <param name="paired_cond|reverse_reads" value="sample.reads_rev.fastq.gz" />
            <param name="matfmt" value="bg2"/>
            <output name="abs_fragments_contacts_weighted_bg2">
                <assert_contents>
                    <has_n_lines n="76"/>
                    <has_n_columns n="7"/>
                </assert_contents>
            </output>
            <output name="fragments_list">
                <assert_contents>
                    <has_n_lines n="17"/>
                    <has_n_columns n="6"/>
                </assert_contents>
            </output>
            <output name="info_contigs" file="info_contigs.txt"/>
            <assert_stderr>
                <has_text text="Contact map generated" />
            </assert_stderr>
        </test>
        <!-- cool format -->
        <test expect_num_outputs="3">
            <param name="genome" value="seq.fa.gz" />
            <param name="paired_cond|paired_select" value="separate"/>
            <param name="paired_cond|forward_reads" value="sample.reads_for.fastq.gz" />
            <param name="paired_cond|reverse_reads" value="sample.reads_rev.fastq.gz" />
            <param name="matfmt" value="cool"/>
            <output name="abs_fragments_contacts_weighted_cool">
                <assert_contents>
                    <has_h5_keys keys="bins,chroms,indexes,pixels"/>
                </assert_contents>
            </output>
            <output name="fragments_list">
                <assert_contents>
                    <has_n_lines n="17"/>
                    <has_n_columns n="6"/>
                </assert_contents>
            </output>
            <output name="info_contigs" file="info_contigs.txt"/>
            <assert_stderr>
                <has_text text="Contact map generated" />
            </assert_stderr>
        </test>
    </tests>
    <help><![CDATA[

hicstuff is a toolkit to generate and manipulate Hi-C matrices. 

The "hicstuff full pipeline" tool generates a Hi-C contact matrix.
Output files can be used with instaGRAAL downstream.

-----------
Input files
-----------
* the fasta genome file
* forward reads
* reverse reads

------------
Output files
------------
* Contact matrix: Sparse matrix file with 3 columns the rows, column and values of nonzero pixels. The first row contains the shape and total number of nonzero pixels in the matrix.
* Fragment ist: Contains genomic coordinates of the matrix bins (row/columns).
* Contig info.txt: Contains chromosome names, theirs length and number of bins.

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