<tool id="gstf_preparation" name="GeneSeqToFamily preparation" version="0.4.3">
    <description>converts data for the workflow</description>
    <requirements>
        <requirement type="package" version="3.7">python</requirement>
    </requirements>
    <command detect_errors="exit_code"><![CDATA[
python '$__tool_directory__/gstf_preparation.py'
#for $q in $queries
    --gff3 '${q.genome}:${q.gff3_input}'
#end for
#if str($json) != 'None'
    #for $v in $json
        --json '$v'
    #end for
#end if
#for $fasta_input in $fasta_inputs
    --fasta '${fasta_input}'
#end for
#if $headers
    --headers $headers
#end if
#if $filter
    --filter $filter
#end if
#if $regions
    --regions '$regions'
    --ff '$filtered_fasta'
#end if
-o '$output_db'
--of '$output_fasta'
    ]]></command>

    <inputs>
        <repeat name="queries" title="GFF3 dataset">
            <param name="gff3_input" type="data" format="gff3" label="GFF3 dataset" />
            <param name="genome" type="text" label="Genome name" help="Genome name without whitespaces or special characters">
                <validator type="empty_field" />
            </param>
        </repeat>
        <param name="json" type="data" format="json" multiple="true" optional="true" label="Gene features in JSON format generated by 'Get features by Ensembl ID' tool" />
        <param name="fasta_inputs" type="data" format="fasta" multiple="true" label="Corresponding CDS datasets in FASTA format" help="Each FASTA header line should start with a transcript id" />
        <param name="filter" type="select" display="radio" label="Which transcripts to keep">
            <option value="canonical" selected="true">Only canonical transcripts (or longest CDS per gene)</option>
            <option value="coding">Only protein-coding transcripts</option>
            <option value="">All transcripts</option>
        </param>

        <param name="headers" type="select" display="radio" label="Change the header line of the FASTA sequences to the following format" help="As required by TreeBest, part of the GeneSeqToFamily workflow, only TranscriptId_species is acceptable format by Aequatus visualisation">
            <option value="TranscriptId_species" selected="true">TranscriptId_species</option>
            <option value="TranscriptID-GeneSymbol_species">TranscriptID-GeneSymbol_species</option>
            <option value="TranscriptID-TranscriptSymbol_species">TranscriptID-TranscriptSymbol_species</option>
            <option value="">Don't change</option>
        </param>
        <param name="regions" type="text" optional="true" label="Comma-separated list of region IDs (e.g. chromosomes or scaffolds) for which FASTA sequences should be filtered out" help="Region IDs are in the `seqid` column for GFF3 and in the `seq_region_name` field in JSON. This is typically used to filter out chromosomes with a non-standard genetic code, like mitochondria, to be analysed separately" />
    </inputs>

    <outputs>
        <data name="output_db" format="sqlite" label="${tool.name} on ${on_string}: SQLite" />
        <data name="output_fasta" format="fasta" label="${tool.name} on ${on_string}: FASTA" />
        <data name="filtered_fasta" format="fasta" label="${tool.name} on ${on_string}: filtered sequences">
            <filter>regions</filter>
        </data>
    </outputs>

    <tests>
        <test expect_num_outputs="2">
            <repeat name="queries">
                <param name="gff3_input" ftype="gff3" value="Caenorhabditis_elegans.WBcel235.87.chromosome.I.shortened.gff3" />
                <param name="genome" value="caenorhabditis_elegans" />
            </repeat>
            <param name="fasta_inputs" ftype="fasta" value="Caenorhabditis_elegans.WBcel235.cds.all.shortened.fa" />
            <param name="filter" value="coding" />
            <param name="headers" value="TranscriptId_species" />

            <output name="output_db" file="test1.sqlite" compare="sim_size" delta="30000" />
            <output name="output_fasta" file="test1.fasta" />
        </test>
        <test expect_num_outputs="2">
            <repeat name="queries">
                <param name="gff3_input" ftype="gff3" value="Caenorhabditis_elegans.WBcel235.87.chromosome.I.shortened.gff3" />
                <param name="genome" value="caenorhabditis_elegans" />
            </repeat>
            <param name="fasta_inputs" ftype="fasta" value="Caenorhabditis_elegans.WBcel235.cds.all.shortened.fa" />
            <param name="filter" value="canonical" />
            <param name="headers" value="TranscriptId_species" />

            <output name="output_db" file="test1.sqlite" compare="sim_size" delta="30000" />
            <output name="output_fasta" file="test1_longest.fasta" />
        </test>
        <test expect_num_outputs="2">
            <param name="json" ftype="gff3" value="gene.json" />
            <param name="fasta_inputs" ftype="fasta" value="CDS.fasta" />
            <param name="filter" value="" />
            <param name="headers" value="" />

            <output name="output_db" file="test4.sqlite" compare="sim_size" delta="30000" />
            <output name="output_fasta" file="CDS.fasta" />
        </test>
        <test expect_num_outputs="2">
            <param name="json" ftype="json" value="gene.json" />
            <param name="fasta_inputs" ftype="fasta" value="CDS.fasta" />
            <param name="filter" value="coding" />
            <param name="headers" value="TranscriptId_species" />

            <output name="output_db" file="test4.sqlite" compare="sim_size" delta="30000" />
            <output name="output_fasta" file="test4.fasta" />
        </test>
        <test>
            <param name="json" ftype="json" value="gene.json" />
            <param name="fasta_inputs" ftype="fasta" value="CDS.fasta" />
            <param name="filter" value="coding" />
            <param name="headers" value="TranscriptId_species" />
            <param name="regions" value="X" />

            <output name="output_db" file="test4.sqlite" compare="sim_size" delta="30000" />
            <output name="output_fasta" file="test5_filtered.fasta" />
            <output name="filtered_fasta" file="test5.ns.fasta" />
        </test>
        <test expect_num_outputs="2">
            <repeat name="queries">
                <param name="gff3_input" ftype="gff3" value="MGP_PahariEiJ_G0008413.1.gff3" />
                <param name="genome" value="mus_pahari" />
            </repeat>
            <param name="fasta_inputs" ftype="fasta" value="Mus_pahari.PAHARI_EIJ_v1.1.cds.all.shortened.fa" />
            <param name="filter" value="canonical" />
            <param name="headers" value="TranscriptId_species" />

            <output name="output_db" file="test6.sqlite" compare="sim_size" delta="30000" />
            <output name="output_fasta" file="test6.fasta" />
        </test>
    </tests>
    <help><![CDATA[
**What it does**

This tool converts a set of GFF3 and/or JSON gene feature information datasets into SQLite format.

It also filters the CDS FASTA datasets to keep only the transcripts present in the gene feature information.

Optionally it can also:
- keep only canonical transcripts (or the longest CDS per gene, if this attribute is not provided)
- remove sequences which are annotated as non protein-coding or whose length is not a multiple of 3
- change the header line of the FASTA sequences to the >TranscriptId_species format (as required by TreeBest, part of the GeneSeqToFamily workflow).

Example GFF3 file::

    scaffold_0  MYZPE13164_Clone_G006_v1.0  gene            44968   69413   .   -   .   ID=MYZPE13164_G006_v1.0_000000030;Name=MYZPE13164_G006_v1.0_000000030;biotype=protein_coding
    scaffold_0  MYZPE13164_Clone_G006_v1.0  mRNA            44968   69413   .   -   .   ID=MYZPE13164_G006_v1.0_000000030.1;Parent=MYZPE13164_G006_v1.0_000000030;Name=MYZPE13164_G006_v1.0_000000030.1;biotype=protein_coding;_AED=0.31
    scaffold_0  MYZPE13164_Clone_G006_v1.0  three_prime_utr 44968   46637   .   -   .   ID=MYZPE13164_G006_v1.0_000000030.1.3utr1;Parent=MYZPE13164_G006_v1.0_000000030.1
    scaffold_0  MYZPE13164_Clone_G006_v1.0  exon            44968   47432   .   -   .   ID=MYZPE13164_G006_v1.0_000000030.1.exon1;Parent=MYZPE13164_G006_v1.0_000000030.1
    scaffold_0  MYZPE13164_Clone_G006_v1.0  CDS             46638   47432   .   -   0   ID=MYZPE13164_G006_v1.0_000000030.1.cds1;Parent=MYZPE13164_G006_v1.0_000000030.1
    scaffold_0  MYZPE13164_Clone_G006_v1.0  exon            53325   53539   .   -   .   ID=MYZPE13164_G006_v1.0_000000030.1.exon2;Parent=MYZPE13164_G006_v1.0_000000030.1
    scaffold_0  MYZPE13164_Clone_G006_v1.0  CDS             53325   53539   .   -   2   ID=MYZPE13164_G006_v1.0_000000030.1.cds2;Parent=MYZPE13164_G006_v1.0_000000030.1
    scaffold_0  MYZPE13164_Clone_G006_v1.0  exon            54614   54719   .   -   .   ID=MYZPE13164_G006_v1.0_000000030.1.exon3;Parent=MYZPE13164_G006_v1.0_000000030.1
    scaffold_0  MYZPE13164_Clone_G006_v1.0  CDS             54614   54719   .   -   0   ID=MYZPE13164_G006_v1.0_000000030.1.cds3;Parent=MYZPE13164_G006_v1.0_000000030.1
    scaffold_0  MYZPE13164_Clone_G006_v1.0  CDS             54852   55106   .   -   0   ID=MYZPE13164_G006_v1.0_000000030.1.cds4;Parent=MYZPE13164_G006_v1.0_000000030.1
    scaffold_0  MYZPE13164_Clone_G006_v1.0  exon            54852   55117   .   -   .   ID=MYZPE13164_G006_v1.0_000000030.1.exon4;Parent=MYZPE13164_G006_v1.0_000000030.1
    scaffold_0  MYZPE13164_Clone_G006_v1.0  five_prime_utr  55107   55117   .   -   .   ID=MYZPE13164_G006_v1.0_000000030.1.5utr1;Parent=MYZPE13164_G006_v1.0_000000030.1
    scaffold_0  MYZPE13164_Clone_G006_v1.0  five_prime_utr  68851   69413   .   -   .   ID=MYZPE13164_G006_v1.0_000000030.1.5utr2;Parent=MYZPE13164_G006_v1.0_000000030.1
    scaffold_0  MYZPE13164_Clone_G006_v1.0  exon            68851   69413   .   -   .   ID=MYZPE13164_G006_v1.0_000000030.1.exon5;Parent=MYZPE13164_G006_v1.0_000000030.1

The following features are parsed: **gene**, **mRNA**, **transcript**, **exon**, **five_prime_utr**, **three_prime_utr** and **CDS**, all other are ignored. Also, **ID** and **Parent** attributes in the 9th column are needed to create relations among features.

.. class:: warningmark

If a value in the **ID** and **Parent** attribute contains a colon, everything up to the first colon will be discarded.
    ]]></help>
    <citations>
    </citations>
</tool>