<tool id="hla_dq" name="HLA-DQ typing" version="1.1">
    <description> Determine possible associated types given BLAST IMGT/HLA annotation</description>
    <command detect_errors="exit_code"><![CDATA[
python '$__tool_directory__/hla_dq.py'
#for $i in $filesA:
    -A '$i.A'
#end for
#for $i in $filesB:
    -B '$i.B'
#end for
#if $column and $column is not None:
    --column $column
#end if
    ]]></command>
    <inputs>
        <!-- use repeats cuz order matters -->
        <repeat name="filesA" title="BLAST results A gene sequences" min="1">
            <param argument="-A" label="BLAST IMGT/HLA hits A gene" type="data" format="tabular"/>
        </repeat>
        <repeat name="filesB" title="BLAST results B gene sequences" min="1">
            <param argument="-B" label="BLAST IMGT/HLA hits B gene" type="data" format="tabular"/>
        </repeat>
        <param argument="--column" label="Column number containing the BLAST IMGT/HLA genotype annotation" type="integer" value="5" min="1" help="Example annotation: HLA:HLA11066 DQA1*01:05:02 768 bp"/>
    </inputs>
    <outputs>
        <data name="serotype_table" format="tabular" from_work_dir="results.tsv" label="${tool.name} on ${on_string}: Serotype table"/>
    </outputs>
    <tests>
        <test><!-- test with real data -->
            <repeat name="filesA">
                <param name="A" value="sideA1.blast.tsv"/>
            </repeat>
            <repeat name="filesA">
                <param name="A" value="sideA2.blast.tsv"/>
            </repeat>
            <repeat name="filesB">
                <param name="B" value="sideB1.blast.tsv"/>
            </repeat>
            <repeat name="filesB">
                <param name="B" value="sideB2.blast.tsv"/>
            </repeat>
            <output name="serotype_table" file="test1_results.tsv"/>
        </test>
        <test><!-- test three files each and data leading to serotypes -->
            <repeat name="filesA">
                <param name="A" value="sideA1.blast.tsv"/>
            </repeat>
            <repeat name="filesA">
                <param name="A" value="sideA2.blast.tsv"/>
            </repeat>
            <repeat name="filesA">
                <param name="A" value="all_sideA3.blast.tsv"/>
            </repeat>
            <repeat name="filesB">
                <param name="B" value="sideB1.blast.tsv"/>
            </repeat>
            <repeat name="filesB">
                <param name="B" value="sideB2.blast.tsv"/>
            </repeat>
             <repeat name="filesB">
                <param name="B" value="all_sideB3.blast.tsv"/>
            </repeat>
            <output name="serotype_table" file="test2_results.tsv"/>
        </test>
    </tests>

    <help><![CDATA[

Given files annotated with BLAST and IMGT/HLA database for HLA-A and HLA-B gene (one or more files each), determine possible associated serotypes.

===== ===== =====
DQA1  DQB1  type
===== ===== =====
02:01 02:02 DQ2.2
03:03 02:02 DQ2.3
05:01 02:01 DQ2.3
03:01 03:02 DQ8
03:02 03:02 DQ8
03:03 03:02 DQ8
===== ===== =====

Example result table:

==== ==== =========
type A1   A2
==== ==== =========
B1   DQ8  None
B2   None DQ2.2;DQ8
==== ==== =========

Where A1, A2 correspond to two annotated input files given for gene A, each representing possible allele sequence

    ]]></help>
</tool>