<tool id="som.py" name="som.py and hap.py" version="@TOOL_VERSION@+galaxy@VERSION_SUFFIX@" profile="@PROFILE@">
    <description>Comparison of somatic callsets based only on chromosome, position, and allele identity</description>
    <macros>
        <token name="@TOOL_VERSION@">0.3.15</token>
        <token name="@PROFILE@">24.1</token>
        <token name="@VERSION_SUFFIX@">0</token>
    </macros>
    <xrefs>
        <xref type="bio.tools">hap.py</xref>
    </xrefs>
    <requirements>
        <requirement type="package" version="@TOOL_VERSION@">hap.py</requirement>
    </requirements>
    <version_command>som.py -h</version_command>
    <command detect_errors="exit_code"><![CDATA[
#if $reference_genome.source == 'history':
    #set $ref_genome = 'reference.fasta'
    ln -s -f '${reference_genome.history_item}' $ref_genome;
    samtools faidx $ref_genome;
#else:
    #set $ref_genome = $reference_genome.index.fields.path
#end if
    export HGREF='$ref_genome' &&
    #if $program_select == 'som.py':
        som.py
    #else:
        hap.py
    #end if
        '$truth'
        '$query'
        -r '$ref_genome'
        -o output
        | sed 's/\s\+/\t/g' | tail -n+2 > results.tsv
    ]]>
    </command>
    <inputs>
        <param name="program_select" type="select" label="Comparison method" help="Select the comparison method to use.">
            <option value="som.py">som.py: Genotype-level comparison</option>
            <option value="hap.py">hap.py: Allele-based comparison (chromosome, position, allele)</option>
        </param>
        <conditional name="reference_genome">
            <param name="source" type="select" label="Source for the reference genome" help="Built-in references were created using default options.">
                <option value="indexed" selected="true">Use a built-in genome</option>
                <option value="history">Use a genome from history</option>
            </param>
            <when value="indexed">
                <param name="index" type="select" label="Select a reference genome" help="If your genome of interest is not listed, contact the Galaxy team.">
                    <options from_data_table="fasta_indexes">
                        <filter type="sort_by" column="2"/>
                        <validator type="no_options" message="No genomes are available for the selected input dataset"/>
                    </options>
                </param>
            </when>
            <when value="history">
                <param name="history_item" type="data" format="fasta" label="Reference genome" help="A reference genome in FASTA format"/>
            </when>
        </conditional>
        <param name="truth" type="data" format="vcf" label="ground truth variant calls"/>
        <param name="query" type="data" format="vcf" label="query variant calls"/>
    </inputs>
    <outputs>
        <data name="results" label="${tool.name} on ${on_string}" format="tabular" from_work_dir="results.tsv"/>
        <data name="sompy_metrics" label="${tool.name} on ${on_string}: som.py metrics" format="json" from_work_dir="output.metrics.json">
            <filter>'som.py' in program_select </filter>
        </data>
        <data name="happy_metrics" label="${tool.name} on ${on_string}: hap.py metrics" format="json" from_work_dir="output.metrics.json.gz">
            <filter>'hap.py' in program_select </filter>
        </data>
        <data name="stats" label="${tool.name} on ${on_string}: som.py comparison stats" format="csv" from_work_dir="output.stats.csv">
            <filter>'som.py' in program_select </filter>
        </data>
        <data name="summary" label="${tool.name} on ${on_string}: hap.py summary" format="csv" from_work_dir="output.summary.csv">
            <filter>'hap.py' in program_select </filter>
        </data>
    </outputs>
    <tests>
        <!-- Testing som.py -->
        <test expect_num_outputs="3">
            <param name="program_select" value="som.py"/>
            <conditional name="reference_genome">
                <param name="source" value="history"/>
                <param name="history_item" value="chr21.fa"/>
            </conditional>
            <param name="truth" value="small_NA12878-Freebayes-chr21.vcf"/>
            <param name="query" value="small_NA12878-GATK3-chr21.vcf"/>
            <output name="results">
                <assert_contents>
                    <has_text text="records"/>
                    <has_n_lines n="4"/>
                </assert_contents>
            </output>
            <output name="stats">
                <assert_contents>
                    <has_line_matching expression="^0,indels,1,1,0,1.*"/>
                    <has_text text="som.py-"/>
                </assert_contents>
            </output>
            <output name="sompy_metrics">
                <assert_contents>
                    <has_json_property_with_text property="name" text="som.py.comparison"/>
                    <has_text text="som.py.comparison"/>
                </assert_contents>
            </output>
        </test>
        <!-- Testing hap.py -->
        <test expect_num_outputs="3">
            <param name="program_select" value="hap.py"/>
            <conditional name="reference_genome">
                <param name="source" value="history"/>
                <param name="history_item" value="reference.fasta"/>
            </conditional>
            <param name="truth" value="hap_py_truth.vcf.gz"/>
            <param name="query" value="hap_py_query.vcf.gz"/>
            <output name="results">
                <assert_contents>
                    <has_n_lines n="6"/>
                </assert_contents>
            </output>
            <output name="summary">
                <assert_contents>
                    <has_n_lines n="5"/>
                    <has_text text="SNP,PASS,5,4,1,6,2,0,0,0,0.8,0.666667,0.0,0.727273,0.25,0.5,0.0,0.0"/>
                </assert_contents>
            </output>
            <output name="happy_metrics" decompress="true">
                <assert_contents>
                    <has_json_property_with_text property="name" text="hap.py.comparison"/>
                    <has_text text="hap.py.comparison"/>
                </assert_contents>
            </output>
        </test>
    </tests>
    <help format="markdown"><![CDATA[

## What this tool does
**hap.py** compares variant callsets (VCFs) against a reference “truth” set to evaluate variant calling accuracy. It performs haplotype-aware comparisons, matching variants by local sequence context rather than simple record alignment. This enables accurate benchmarking even when complex or multi-nucleotide variants are represented differently.
---

## Inputs
- **Truth VCF:** High-confidence reference variants.  
- **Query VCF:** Variants to evaluate.  
- **Reference FASTA:** Genome reference used for both callsets.  
---

## Outputs
- **Summary CSV / JSON:** Precision, recall, and F1 metrics.  
- **ROC and stratified metrics:** Optional, for detailed performance breakdowns.  
- **Annotated VCFs:** Optional lists of matched and unmatched variants.
---

## Notes
- Use **hap.py** for haplotype-level benchmarking.  
- Use **som.py** for simpler allele-based comparisons (chromosome, position, allele).  
    ]]></help>
    <citations>
        <citation type="bibtex">@misc{illumina_happy,
  title        = {hap.py},
  author       = {{Illumina}},
  year         = {2021},
  howpublished = {\url{https://github.com/Illumina/hap.py}},
  note         = {Accessed: 2025-10-19}
}</citation>
    </citations>
</tool>