<tool id="read_it_and_keep" name="Read It and Keep" version="@TOOL_VERSION@+galaxy0" profile="20.09">
    <macros>
        <token name="@INPUT_FORMATS@">fasta,fastq,fasta.gz,fastq.gz</token>
        <token name="@TOOL_VERSION@">0.2.2</token>
    </macros>
    <requirements>
        <requirement type="package" version="@TOOL_VERSION@">read-it-and-keep</requirement>
        <requirement type="package" version="3.10">python</requirement>
    </requirements>
    <version_command><![CDATA[readItAndKeep -V]]></version_command>
    <command detect_errors="exit_code"><![CDATA[
        #if $ref_source.source == "history"
            ln -s '$ref_source.ref_fasta' ref${trim_reference}.fasta &&
        #elif $ref_source.source == "builtin"
            ln -s '$ref_source.ref_fasta_builtin.path' ref${trim_reference}.fasta &&
        #end if
        #if $trim_reference
            python '$__tool_directory__/trim_reference.py' ref${trim_reference}.fasta ref.fasta &&
        #end if
        #if $reads.read_type == "paired"
            #set $ext_out1 = 'fasta' if $reads.read1.ext in ['fasta', 'fasta.gz'] else 'fastq'
            #set $ext_out2 = 'fasta' if $reads.read1.ext in ['fasta', 'fasta.gz'] else 'fastq'
            ln -s '$reads.read1' read1 &&
            ln -s '$reads.read2' read2 &&
        #elif $reads.read_type == 'paired_collection'
        #set $ext_out1 = 'fasta' if $reads.paired_reads.forward.ext in ['fasta', 'fasta.gz'] else 'fastq'
            #set $ext_out2 = 'fasta' if $reads.paired_reads.reverse.ext in ['fasta', 'fasta.gz'] else 'fastq'
            ln -s '$reads.paired_reads.forward' read1 &&
            ln -s '$reads.paired_reads.reverse' read2 &&
        #elif $reads.read_type == 'single'
            #set $ext_out1 = 'fasta' if $reads.read1.ext in ['fasta', 'fasta.gz'] else 'fastq'
            ln -s '$reads.read1' read1 &&
        #end if
        readItAndKeep
            --tech $sequencing_tech
            --ref_fasta ref.fasta
            --min_map_length $adv.min_map_length
            --min_map_length_pc $adv.min_map_length_pc
            $adv.enumerate_names
            --reads1 read1
            #if $reads.read_type != "single"
                --reads2 read2
            #end if
            -o output
            #if $reads.read_type == 'single':
                && mv output.reads.${ext_out1}.gz output.reads_1.${ext_out1}.gz
            #end if
            #if $reads.read_type == 'paired_collection':
                #if not $reads.paired_reads.forward.ext.endswith('.gz'):
                    && gunzip ./output.reads_1.${ext_out1}.gz
                    && gunzip ./output.reads_2.${ext_out2}.gz
                #end if
            #elif not $reads.read1.ext.endswith('.gz'):
                && gunzip ./output.reads_1.${ext_out1}.gz
                #if $reads.read_type != 'single' and not $reads.read2.ext.endswith('.gz'):
                    && gunzip ./output.reads_2.${ext_out2}.gz
                #end if
            #end if
    ]]></command>
    <inputs>
        <conditional name="reads">
            <param type="select" label="Read type" name="read_type">
                <option value="paired" selected="true">Paired end</option>
                <option value="paired_collection">Paired collection</option>
                <option value="single">Single ended</option>
            </param>
            <when value="paired">
                <param type="data" format="@INPUT_FORMATS@" name="read1" label="Read1" />
                <param type="data" format="@INPUT_FORMATS@" name="read2" label="Read2" />
            </when>
            <when value="paired_collection">
                <param type="data_collection" collection_type="paired" format="@INPUT_FORMATS@" name="paired_reads" label="Reads" />
            </when>
            <when value="single">
                <param type="data" format="@INPUT_FORMATS@" name="read1" label="Read1" />
            </when>
        </conditional>
        <conditional name="ref_source">
            <param type="select" label="Reference genome source" name="source">
                <option value="history" selected="true">History</option>
                <option value="builtin">Built-in</option>
            </param>
            <when value="history">
                <param type="data" format="fasta" name="ref_fasta" label="Reference genome" />
            </when>
            <when value="builtin">
                <param type="select" name="ref_fasta_builtin" label="Reference genome">
                    <options from_data_table="all_fasta" />
                </param>
            </when>
        </conditional>
        <param type="boolean" name="trim_reference" label="Trim trailing As from the reference sequence" checked="true" truevalue="_untrimmed" falsevalue="" help="Remove all As at the end of the reference sequence to ensure that the reference has no poly-A tail (see Note in the general help section below)" />
        <param type="select" argument="--tech" name="sequencing_tech" label="Sequencing technology">
            <option value="illumina">Illumina</option>
            <option value="ont">Oxford Nanopore</option>
        </param>
        <section name="adv" title="Advanced options">
            <param argument="--min_map_length" type="integer" min="0" value="50" label="Shortest match required to keep a read (in bp)" />
            <param argument="--min_map_length_pc" type="float" min="0.0" max="100.0" value="50.0" label="Minimum length of match required to keep a read (as percentage of read length" />
            <param name="enumerate_names" type="boolean" truevalue="--enumerate_names" falsevalue="" label="Rename the reads 1,2,3,... (for paired reads, will also add /1 or /2 to the end of names)" />
        </section>
    </inputs>
    <outputs>
        <data name="output_reads1" format_source="read1" label="Filtered reads ${on_string} - reads1" from_work_dir="output.reads_1.fast*">
            <filter>reads["read_type"] == "single" or reads["read_type"] == "paired"</filter>
        </data>
        <data name="output_reads2" format_source="read2" label="Filtered reads ${on_string} - reads2" from_work_dir="output.reads_2.fast*">
            <filter>reads["read_type"] == "paired"</filter>
        </data>
        <collection type="paired" name="output_collection" format_source="paired_reads" label="Filtered reads ${on_string}">
            <filter>reads["read_type"] == "paired_collection"</filter>
            <data name="forward" from_work_dir="output.reads_1.fast*" />
            <data name="reverse" from_work_dir="output.reads_2.fast*" />
        </collection>
    </outputs>
    <tests>
        <!--Testing uncompressed fastq input-->
        <test expect_num_outputs="2">
            <conditional name="reads">
                <param name="read_type" value="paired" />
                <param name="read1" value="test1.fastq.gz" ftype="fastqsanger" />
                <param name="read2" value="test2.fastq.gz" ftype="fastqsanger" />
            </conditional>
            <conditional name="ref_source">
                <param name="source" value="history" />
                <param name="ref_fasta" value="reference.fasta" ftype="fasta" />
            </conditional>
            <param name="sequencing_tech" value="illumina" />
            <output name="output_reads1" file="output.test1.reads_1.fastq" ftype="fastqsanger"/>
            <output name="output_reads2" file="output.test1.reads_2.fastq" ftype="fastqsanger"/>
        </test>
        <test expect_num_outputs="3">
            <conditional name="reads">
                <param name="read_type" value="paired_collection" />
                <param name="paired_reads">
                    <collection type="paired">
                        <element name="forward" value="test1.fastq.gz" ftype="fastqsanger" />
                        <element name="reverse" value="test2.fastq.gz" ftype="fastqsanger" />
                    </collection>
                </param>
            </conditional>
            <conditional name="ref_source">
                <param name="source" value="history" />
                <param name="ref_fasta" value="reference.fasta" ftype="fasta" />
            </conditional>
            <param name="sequencing_tech" value="illumina" />
            <output_collection name="output_collection" type="paired">
                <element name="forward" file="output.test1.reads_1.fastq" ftype="fastqsanger"/>
                <element name="reverse" file="output.test1.reads_2.fastq" ftype="fastqsanger"/>
            </output_collection>
        </test>
        <test expect_num_outputs="1">
            <conditional name="reads">
                <param name="read_type" value="single" />
                <param name="read1" value="test3.fastq.gz" ftype="fastqsanger"/>
            </conditional>
            <conditional name="ref_source">
                <param name="source" value="history" />
                <param name="ref_fasta" value="reference.fasta" ftype="fasta" />
            </conditional>
            <param name="sequencing_tech" value="ont" />
            <param name="enumerate_names" value="True" />
            <output name="output_reads1" file="output.test3.reads_1.fastq" ftype="fastqsanger"/>
        </test>
        <!--Testing uncompressed fasta input-->
        <test expect_num_outputs="2">
            <conditional name="reads">
                <param name="read_type" value="paired" />
                <param name="read1" value="test1.fasta.gz" ftype="fasta" />
                <param name="read2" value="test2.fasta.gz" ftype="fasta" />
            </conditional>
            <conditional name="ref_source">
                <param name="source" value="history" />
                <param name="ref_fasta" value="reference.fasta" ftype="fasta" />
            </conditional>
            <param name="sequencing_tech" value="illumina" />
            <output name="output_reads1" file="output.test1.reads_1.fasta" ftype="fasta"/>
            <output name="output_reads2" file="output.test1.reads_2.fasta" ftype="fasta"/>
        </test>
        <test expect_num_outputs="3">
            <conditional name="reads">
                <param name="read_type" value="paired_collection" />
                <param name="paired_reads">
                    <collection type="paired">
                        <element name="forward" value="test1.fasta.gz" ftype="fasta" />
                        <element name="reverse" value="test2.fasta.gz" ftype="fasta" />
                    </collection>
                </param>
            </conditional>
            <conditional name="ref_source">
                <param name="source" value="history" />
                <param name="ref_fasta" value="reference.fasta" ftype="fasta" />
            </conditional>
            <param name="sequencing_tech" value="illumina" />
            <output_collection name="output_collection" type="paired">
                <element name="forward" file="output.test1.reads_1.fasta" ftype="fasta"/>
                <element name="reverse" file="output.test1.reads_2.fasta" ftype="fasta"/>
            </output_collection>
        </test>
        <test expect_num_outputs="1">
            <conditional name="reads">
                <param name="read_type" value="single" />
                <param name="read1" value="test3.fasta.gz" ftype="fasta"/>
            </conditional>
            <conditional name="ref_source">
                <param name="source" value="history" />
                <param name="ref_fasta" value="reference.fasta" ftype="fasta" />
            </conditional>
            <param name="sequencing_tech" value="ont" />
            <param name="enumerate_names" value="True" />
            <output name="output_reads1" file="output.test3.reads_1.fasta" ftype="fasta"/>
        </test>
        <!--Testing compressed fasta input-->
        <test expect_num_outputs="2">
            <conditional name="reads">
                <param name="read_type" value="paired" />
                <param name="read1" value="test1.fasta.gz" ftype="fasta.gz" />
                <param name="read2" value="test2.fasta.gz" ftype="fasta.gz" />
            </conditional>
            <conditional name="ref_source">
                <param name="source" value="history" />
                <param name="ref_fasta" value="reference.fasta" ftype="fasta" />
            </conditional>
            <param name="sequencing_tech" value="illumina" />
            <output name="output_reads1" decompress="true" file="output.test1.reads_1.fasta" ftype="fasta.gz"/>
            <output name="output_reads2" decompress="true" file="output.test1.reads_2.fasta" ftype="fasta.gz"/>
        </test>
        <test expect_num_outputs="3">
            <conditional name="reads">
                <param name="read_type" value="paired_collection" />
                <param name="paired_reads">
                    <collection type="paired">
                        <element name="forward" value="test1.fasta.gz" ftype="fasta.gz" />
                        <element name="reverse" value="test2.fasta.gz" ftype="fasta.gz" />
                    </collection>
                </param>
            </conditional>
            <conditional name="ref_source">
                <param name="source" value="history" />
                <param name="ref_fasta" value="reference.fasta" ftype="fasta" />
            </conditional>
            <param name="sequencing_tech" value="illumina" />
            <output_collection name="output_collection" type="paired">
                <element name="forward" ftype="fasta.gz">
                    <assert_contents>
                        <has_size value="680" delta="50" />
                    </assert_contents>
                </element>
                <element name="reverse" ftype="fasta.gz">
                    <assert_contents>
                        <has_size value="670" delta="50" />
                    </assert_contents>
                </element>
            </output_collection>
        </test>
        <test expect_num_outputs="1">
            <conditional name="reads">
                <param name="read_type" value="single" />
                <param name="read1" value="test3.fasta.gz" ftype="fasta.gz"/>
            </conditional>
            <conditional name="ref_source">
                <param name="source" value="history" />
                <param name="ref_fasta" value="reference.fasta" ftype="fasta" />
            </conditional>
            <param name="sequencing_tech" value="ont" />
            <param name="enumerate_names" value="True" />
            <output name="output_reads1" decompress="true" file="output.test3.reads_1.fasta" ftype="fasta.gz"/>
        </test>
        <!--Testing compressed fastq input-->
        <test expect_num_outputs="2">
            <conditional name="reads">
                <param name="read_type" value="paired" />
                <param name="read1" value="test1.fastq.gz" ftype="fastqsanger.gz" />
                <param name="read2" value="test2.fastq.gz" ftype="fastqsanger.gz" />
            </conditional>
            <conditional name="ref_source">
                <param name="source" value="history" />
                <param name="ref_fasta" value="reference.fasta" ftype="fasta" />
            </conditional>
            <param name="sequencing_tech" value="illumina" />
            <output name="output_reads1" decompress="true" file="output.test1.reads_1.fastq" ftype="fastqsanger.gz"/>
            <output name="output_reads2" decompress="true" file="output.test1.reads_2.fastq" ftype="fastqsanger.gz"/>
        </test>
        <test expect_num_outputs="3">
            <conditional name="reads">
                <param name="read_type" value="paired_collection" />
                <param name="paired_reads">
                    <collection type="paired">
                        <element name="forward" value="test1.fastq.gz" ftype="fastqsanger.gz" />
                        <element name="reverse" value="test2.fastq.gz" ftype="fastqsanger.gz" />
                    </collection>
                </param>
            </conditional>
            <conditional name="ref_source">
                <param name="source" value="history" />
                <param name="ref_fasta" value="reference.fasta" ftype="fasta" />
            </conditional>
            <param name="sequencing_tech" value="illumina" />
            <output_collection name="output_collection" type="paired">
                <element name="forward" ftype="fastqsanger.gz">
                    <assert_contents>
                        <has_size value="995" delta="50" />
                    </assert_contents>
                </element>
                <element name="reverse" ftype="fastqsanger.gz">
                    <assert_contents>
                        <has_size value="960" delta="50" />
                    </assert_contents>
                </element>
            </output_collection>
        </test>
        <test expect_num_outputs="1">
            <conditional name="reads">
                <param name="read_type" value="single" />
                <param name="read1" value="test3.fastq.gz" ftype="fastqsanger.gz"/>
            </conditional>
            <conditional name="ref_source">
                <param name="source" value="history" />
                <param name="ref_fasta" value="reference.fasta" ftype="fasta" />
            </conditional>
            <param name="sequencing_tech" value="ont" />
            <param name="enumerate_names" value="True" />
            <output name="output_reads1" decompress="true" file="output.test3.reads_1.fastq" ftype="fastqsanger.gz"/>
        </test>
    </tests>
    <help><![CDATA[
ReadItAndKeep
-------------

ReadItAndKeep is a tool for filtering viral sequence data to remove host reads, developed for cleaning
SARS-CoV-2 sequencing data. It maps reads against the SARS-CoV-2 viral genome (with the poly-A tail removed)
and only keeps those that map well.

**Note**: If the reference genome supplied contains a poly-A tail, reads that contain part of a poly-A tail will map
to the refence, no matter what species they originate from. If you are not sure if the reference you are using has had
trailing A's trimmed, enable the `Trim trailing As` option.

Input can be either Illumina or Oxford Nanopore reads.
    ]]></help>
    <citations>
        <citation type="doi">10.1101/2022.01.21.477194</citation>
    </citations>
</tool>