<tool id="vsnp_determine_ref_from_data" name="vSNP: determine reference" version="@TOOL_VERSION@+galaxy@VERSION_SUFFIX@" profile="@PROFILE@">
    <description>from input data</description>
    <macros>
        <import>macros.xml</import>
    </macros>
    <requirements>
        <expand macro="biopython_requirement"/>
        <expand macro="pyyaml_requirement"/>
    </requirements>
    <command detect_errors="exit_code"><![CDATA[
#import re
#set gzipped = 'false'
#set input_type = $input_type_cond.input_type

#if $input_type in ["single", "pair"]:
    #set read1 = $input_type_cond.read1
    #set read1_identifier = re.sub('[^\s\w\-]', '_', str($read1.element_identifier))
    ln -s '${read1}' '${read1_identifier}' &&
    #if $input_type == "pair":
        #set read2 = $input_type_cond.read2
        #set read2_identifier = re.sub('[^\s\w\-]', '_', str($read2.element_identifier))
        ln -s '${read2}' '${read2_identifier}' &&
    #else:
        #set read2 = None 
    #end if
#else:
    #set read1 = $input_type_cond.reads_collection['forward']
    #set read1_identifier = re.sub('[^\s\w\-]', '_', str($read1.name))
    ln -s '${read1}' '${read1_identifier}' &&
    #set read2 = $input_type_cond.reads_collection['reverse']
    #set read2_identifier = re.sub('[^\s\w\-]', '_', str($read2.name))
    ln -s '${read2}' '${read2_identifier}' &&
#end if

python '$__tool_directory__/vsnp_determine_ref_from_data.py'
    --read1 '${read1_identifier}'
    #if $read2 is not None
      --read2 '${read2_identifier}'
    #end if
    --output_dbkey '$output_dbkey'
    --output_metrics '$output_metrics'
#if $read1.is_of_type('fastqsanger.gz'):
    --gzipped
#end if
#set $dnaprint_fields = $__app__.tool_data_tables['vsnp_dnaprints'].get_fields()
#for $i in $dnaprint_fields:
    --dnaprint_fields '${i[0]}' '${i[2]}'
#end for
]]></command>
    <inputs>
        <conditional name="input_type_cond">
            <param name="input_type" type="select" label="Choose the category of the files to be analyzed">
                <option value="single" selected="true">Single files</option>
		<option value="paired">Paired reads</option>
		<option value="pair">Paired reads in separate data sets</option>
            </param>
            <when value="single">
                <param name="read1" type="data" format="fastqsanger.gz,fastqsanger" label="Read1 fastq file"/>
            </when>
            <when value="paired">
                <param name="reads_collection" type="data_collection" format="fastqsanger,fastqsanger.gz" collection_type="paired" label="Collection of fastqsanger paired read files"/>
            </when>
            <when value="pair">
                <param name="read1" type="data" format="fastqsanger.gz,fastqsanger" label="Read1 fastq file"/>
                <param name="read2" type="data" format="fastqsanger.gz,fastqsanger" label="Read2 fastq file"/>
            </when>
        </conditional>
    </inputs>
    <outputs>
        <data name="output_dbkey" format="txt" label="${tool.name} on ${on_string} (dbkey)"/>
        <data name="output_metrics" format="txt" label="${tool.name} on ${on_string} (metrics)"/>
    </outputs>
    <tests>
        <!-- 1 single read -->
        <test expect_num_outputs="2">
            <param name="input_type" value="single"/>
            <param name="read1" value="Mcap_Deer_DE_SRR650221.fastq.gz" ftype="fastqsanger.gz"/>
            <output name="output_dbkey" file="output_dbkey.txt" ftype="txt"/>
            <output name="output_metrics" file="output_metrics.txt" ftype="txt"/>
        </test>
        <!-- 1 set of paired reads -->
        <test expect_num_outputs="2">
            <param name="input_type" value="pair"/>
            <param name="read1" value="CMC_20E1_R1.fastq.gz" ftype="fastqsanger.gz"/>
            <param name="read2" value="CMC_20E1_R2.fastq.gz" ftype="fastqsanger.gz"/>
            <output name="output_dbkey" file="paired_dbkey.txt" ftype="txt"/>
            <output name="output_metrics" file="paired_metrics.txt" ftype="txt"/>
        </test>
        <!-- A collection of paired reads -->
        <test expect_num_outputs="2">
            <param name="input_type" value="paired"/>
            <param name="reads_collection">
                <collection type="paired">
                    <element name="forward" value="CMC_20E1_R1.fastq.gz" ftype="fastqsanger.gz"/>
                    <element name="reverse" value="CMC_20E1_R2.fastq.gz" ftype="fastqsanger.gz"/>
                </collection>
            </param>
            <output name="output_dbkey" file="paired_dbkey.txt" ftype="txt"/>
            <output name="output_metrics" file="paired_collection_metrics.txt" ftype="txt"/>
        </test>
    </tests>
    <help>
**What it does**

Accepts a single fastqsanger read, a set of paired reads, or a collection of single or paired reads (bacterial samples) and
inspects the data to discover the best reference genome for aligning the reads.

The information needed to discover the best reference is maintained by the USDA in this repository_.  References are curreently

.. _repository:  https://github.com/USDA-VS/vSNP_reference_options

limited to TB complex, paraTB, and Brucella, but information for additional references will be added.  The information for each
reference is a string consisting of zeros and ones, compiled by USDA researchers, which we call a "DNA print".   These strings
are maintained in yaml files for use in Galaxy, and are installed via the **vSNP DNAprints data manager** tool.

This tool creates an in-memory dictionary of these DNA print strings for matching with a string generated by inspecting the
input sample data.  During inspection, this tool accrues sequence counts for supported species, ultimately generating a string
consisting of zeros and ones based on the counts, (i.e., a DNA print).  This string is then compared to the strings contained
in the in-memory dictionary of DNA prints to find a match.

The strings in the in-memory dictionary are each associated with a Galaxy "dbkey" (i.e., genome build), so when a match is found,
the associated "dbkey" is passed to a mapper (e.g., **Map with BWA-MEM**), typically within a workflow via an expression tool,
to align the reads to the associated reference.

This tool produces 2 text files, a "dbkey" file that contains the dbkey string and a "metrics" file that provides information
about the sequence counts that were discovered in the input sample data that produced the "DNA print" string.

This tool is important for samples containing bacterial species because many of the samples have a "mixed bag" of species,
and discovering the primary species is critical.  DNA print matching is currently supported for the following genomes.

 * Mycobacterium bovis AF2122/97
 * Brucella abortus bv. 1 str. 9-941
 * Brucella abortus strain BER
 * Brucella canis ATCC 23365
 * Brucella ceti TE10759-12
 * Brucella melitensis bv. 1 str. 16M
 * Brucella melitensis bv. 3 str. Ether
 * Brucella melitensis BwIM_SOM_36b
 * Brucella melitensis ATCC 23457
 * Brucella ovis ATCC 25840
 * Brucella suis 1330
 * Mycobacterium tuberculosis H37Rv
 * Mycobacterium avium subsp. paratuberculosis strain Telford
 * Mycobacterium avium subsp. paratuberculosis K-10
 * Brucella suis ATCC 23445
 * Brucella suis bv. 3 str. 686

**Required Options**

 * **Choose the category of the files to be analyzed** - select "Single files" or "Collection of files", then select the appropriate history items (single or paired fastqsanger reads or a collection of fastqsanger reads) based on the selected option.
    </help>
    <expand macro="citations"/>
</tool>