<tool id="Kraken2Tax" name="Convert Kraken" version="1.2">
  <description>data to Galaxy taxonomy representation</description>
    <requirements>
        <requirement type="package" version="4.1.0">gawk</requirement>
        <requirement type="package" version="1.0.0">gb_taxonomy_tools</requirement>
    </requirements>
  <command>
<![CDATA[
  awk '{ print \$${read_name}, \$${tax_id} }' OFS="\t" "${input}" | taxonomy-reader "${ncbi_taxonomy.fields.path}/names.dmp" "${ncbi_taxonomy.fields.path}/nodes.dmp" 1 > "${out_file}"
]]>
  </command>
  <inputs>
    <param format="tabular" name="input" type="data" label="Choose dataset to convert"/>
    <param label="Select a taxonomy database" name="ncbi_taxonomy" type="select">
        <options from_data_table="ncbi_taxonomy">
            <validator message="No built-in databases are available" type="no_options" />
        </options>
    </param>
    <param name="read_name" label="Read name" type="data_column" data_ref="input" value="2" help="Select column containing read names"/>
    <param name="tax_id" label="Taxonomy ID field" type="data_column" data_ref="input" numerical="True" value="3" help="Select column containing taxonomy ID"/>
  </inputs>
  <outputs>
    <data format="taxonomy" name="out_file" />
  </outputs>
  <tests>
    <test>
      <param name="input" ftype="tabular" value="kraken2tax.txt"/>
      <param name="read_name" value="2"/>
      <param name="tax_id" value="3"/>
      <output name="out_file" file="kraken2tax-test1.txt"/>
    </test>
  </tests>
  <help>

.. class:: infomark

Use *Filter and Sort->Filter* to restrict output of this tool to desired taxonomic ranks. You can also use *Text Manipulation->Cut* to remove unwanted columns from the output.

------

**What it does**

This tool is designed to translate results of the Kraken metagenomic classifier (see citations below) to the full representation of NCBI taxonomy. It does so by using Taxonomic ID field provided by Kraken. The output of this tool can be directly visualized by the Krona tool. It is based on `gb_taxonomy_tools` developed by https://github.com/spond.

-------

**Example**

Suppose you have Kraken output that looks like this (here the second field is the name of a sequencing read and the third is the taxonomic ID)::

  C   Read_1   9606  465   Q:1

and you want to obtain the full taxonomic representation for this read. Setting **Read name** and **Taxonomy ID field** parameters to **2** and **3**, respectively, will produce the following output (you may need to scroll sideways to see the entire line)::

  1      2    3    4         5       6 7 8        9        10            11       12 13               14       15         16          17        18  19  20 21  22  23           24
  Read_1 9606 root Eukaryota Metazoa n n Chordata Craniata Gnathostomata Mammalia n  Euarchontoglires Primates Haplorrhini Hominoidea Hominidae n   n   n  Homo n  Homo sapiens n

In other words the tool printed *Read name*, *Taxonomy ID field*, and appended 22 columns containing taxonomic ranks from Superkingdom to Subspecies. Below is a formal definition of the output columns::

    Column Definition
   ------- -------------------------------------------------
         1 Name (specified by 'Read name' dropdown)
         2 taxID   (specified by 'Taxonomy ID field' dropdown)
         3 root
         4 superkingdom
         5 kingdom
         6 subkingdom
         7 superphylum
         8 phylum
         9 subphylum
        10 superclass
        11 class
        12 subclass
        13 superorder
        14 order
        15 suborder
        16 superfamily
        17 family
        18 subfamily
        19 tribe
        20 subtribe
        21 genus
        22 subgenus
        23 species
        24 subspecies

------

.. class:: warningmark

**Why do I have these "n" things?** 

Be aware that the NCBI taxonomy (ftp://ftp.ncbi.nih.gov/pub/taxonomy/) this tool relies upon is incomplete.  This means that for many species one or more ranks are absent and represented as "**n**". In the above example *subkingdom*, *superphylum* etc. are missing.


  </help>
  <citations>
    <citation type="doi">10.1186/gb-2014-15-3-r46</citation>
    <citation type="doi">10.1101/gr.094508.109</citation>
  </citations>
</tool>