<?xml version="1.0"?>
<tool id="bigbed_to_fa" name="bigBedToFa" version="1.0">
    <description>Retrieve sequences associated with bigBed features</description>

    <macros>
        <import>ucsc_macros.xml</import>
    </macros>

    <expand macro="requirements_twobit">
        <requirement type="package" version="340">ucsc_bigbed</requirement>
    </expand>

    <command detect_errors="exit_code">
<![CDATA[
    #set no_mask = ""
    #if str($masking_option) == "upper":
        #set no_mask = "-noMask"
    #end if

    ${__tool_directory__}/bigbed_to_standard_bed.pl -i "${bigbed_input}" |
        twoBitToFa ${no_mask} -bed=stdin "${twobit_input}" stdout

    ## Add extra option to change soft masked bases to N's
    #if str($masking_option) == "hard":
        | awk '{ if (/^>/) { print } else { gsub(/[acgt]/, "N"); print } }'
    #end if

        > "${fasta_output}"
]]>
    </command>
    <inputs>
        <param name="twobit_input" type="data" format="twobit" label="twoBit input file" />

        <param name="bigbed_input" type="data" format="bigbed" label="bigBed input file" />

        <param name="masking_option" type="select"
            label="Repeat masking option"
            help="Specify how repeats within the twoBit file should be represented">

            <option value="upper">Show repeats in uppercase (no masking)</option>

            <option value="lower" selected="true">Show repeats in lowercase (soft masking)</option>

            <option value="hard">Show repeats as N's (hard masking)</option>
        </param>

    </inputs>
    <outputs>
        <data name="fasta_output" format="fasta" />
    </outputs>
    <tests>
        <test>
            <!-- Test bigBedToFa with bed4 and default (soft) masking -->
            <param name="twobit_input" value="contigs.out.2bit" ftype="twobit" />
            <param name="bigbed_input" value="contigs.trf_4.bb" ftype="bigbed" />
            <output name="fasta_output" file="contigs.trf_4.soft.fa" />
        </test>
        <test>
            <!-- Test bigBedToFa with bed4 and no masking -->
            <param name="twobit_input" value="contigs.out.2bit" ftype="twobit" />
            <param name="bigbed_input" value="contigs.trf_4.bb" ftype="bigbed" />
            <param name="masking_option" value="upper" />
            <output name="fasta_output" file="contigs.trf_4.nomask.fa" />
        </test>
        <test>
            <!-- Test bigBedToFa with bed4 and hard masking -->
            <param name="twobit_input" value="contigs.out.2bit" ftype="twobit" />
            <param name="bigbed_input" value="contigs.trf_4.bb" ftype="bigbed" />
            <param name="masking_option" value="hard" />
            <output name="fasta_output" file="contigs.trf_4.hard.fa" />
        </test>
        <test>
            <!-- Test bigBedToFa with bed4+12 with default masking -->
            <param name="twobit_input" value="contigs.out.2bit" ftype="twobit" />
            <param name="bigbed_input" value="contigs.trf_4_12.bb" ftype="bigbed" />
            <output name="fasta_output" file="contigs.trf_4_12.fa" />
        </test>
        <test>
            <!-- Test bigBedToFa with bed6+2 with default masking -->
            <param name="twobit_input" value="contigs.out.2bit" ftype="twobit" />
            <param name="bigbed_input" value="contigs.models_6_2.bb" ftype="bigbed" />
            <output name="fasta_output" file="contigs.models_6_2.fa" />
        </test>
        <test>
            <!-- Test bigBedToFa with bed12+2 with default masking -->
            <param name="twobit_input" value="contigs.out.2bit" ftype="twobit" />
            <param name="bigbed_input" value="contigs.models_12_2.bb" ftype="bigbed" />
            <output name="fasta_output" file="contigs.models_12_2.fa" />
        </test>
    </tests>
    <help>
<![CDATA[
**What it does**

bigBedToFa retrieves sequences associated with the features in a
`bigBed <https://genome.ucsc.edu/goldenpath/help/bigBed.html>`_ file
from a `twoBit Sequence Archive <https://genome.ucsc.edu/goldenpath/help/twoBit.html>`_.

.. class:: infomark

**Extracted sequences:**

For files with `12 standard BED columns <https://genome.ucsc.edu/FAQ/FAQformat.html#format1>`_,
this program will concatenate the sequences associated with each block (e.g., exons), and
exclude the sequences between adjacent blocks (e.g., introns).

    ]]></help>

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