annotate pileup_parser.xml @ 2:85bedbea8a12 draft default tip

planemo upload for repository https://github.com/galaxyproject/tools-devteam/tree/master/tools/pileup_parser commit ab627176cd4f6efe6d1fe4b85baa679aaa651eb1

<tool id="pileup_parser" name="Filter pileup" version="1.0.2">
  <description>on coverage and SNPs</description>
  <requirements>
        <requirement type="package" version="5.22.0">perl</requirement>
   </requirements>
  <command interpreter="perl">
    #if   $pileup_type.type_select == "six"    #pileup_parser.pl $input "3" "5" "6" "4" $qv_cutoff $cvrg_cutoff $snps_only $interval "2" $out_file1 $diff $qc_base
    #elif $pileup_type.type_select == "ten"    #pileup_parser.pl $input "3" "9" "10" "8" $qv_cutoff $cvrg_cutoff $snps_only $interval "2" $out_file1 $diff $qc_base
    #elif $pileup_type.type_select == "manual" #pileup_parser.pl $input $pileup_type.ref_base_column $pileup_type.read_bases_column $pileup_type.read_qv_column $pileup_type.cvrg_column $qv_cutoff $cvrg_cutoff $snps_only $interval $pileup_type.coord_column $out_file1 $diff $qc_base
    #end if#
  </command>
  <inputs>
    <param name="input" type="data" format="pileup" label="Select dataset"/>
    <conditional name="pileup_type">
        <param name="type_select" type="select" label="which contains" help="See &quot;Types of pileup datasets&quot; below for examples">
            <option value="six" selected="true">Pileup with six columns (simple)</option>
            <option value="ten">Pileup with ten columns (with consensus)</option>
            <option value="manual">Set columns manually</option>
        </param>
        <when value="manual">
            <param name="ref_base_column" label="Select column with reference base" type="data_column" numerical="false" data_ref="input" />
            <param name="read_bases_column" label="Select column with read bases" type="data_column" numerical="false" data_ref="input" help="something like this: ..,a.."/>
            <param name="read_qv_column" label="Select column with base qualities" type="data_column" numerical="false" data_ref="input" help="something like this: IIIGIAI"/>
            <param name="cvrg_column" label="Select column with coverage" type="data_column" numerical="true" data_ref="input" />
            <param name="coord_column" label="Select coordinate column" type="data_column" numerical="true" data_ref="input" />
        </when>
        <when value="six">
        </when>
        <when value="ten">
        </when>
    </conditional>
    <param name="qv_cutoff" label="Do not consider read bases with quality lower than" type="integer" value="20" help="No variants with quality below this value will be reported"/>
    <param name="cvrg_cutoff" label="Do not report positions with coverage lower than" type="integer" value="3" help="Pileup lines with coverage lower than this value will be skipped"/>
    <param name="snps_only" label="Only report variants?" type="select" help="See &quot;Examples 1 and 2&quot; below for explanation">
        <option value="No">No</option>
        <option value="Yes" selected="true">Yes</option>
    </param>
    <param name="interval" label="Convert coordinates to intervals?" type="select" help="See &quot;Output format&quot; below for explanation">
        <option value="No" selected="true">No</option>
        <option value="Yes">Yes</option>
    </param>
    <param name="diff" label="Print total number of differences?" type="select" help="See &quot;Example 3&quot; below for explanation">
        <option value="No" selected="true">No</option>
        <option value="Yes">Yes</option>
    </param>
    <param name="qc_base" label="Print quality and base string?" type="select" help="See &quot;Example 4&quot; below for explanation">
        <option value="No">No</option>
        <option value="Yes" selected="true">Yes</option>
    </param>
  </inputs>
  <outputs>
    <data format="tabular" name="out_file1">
        <change_format>
            <when input="interval" value="Yes" format="interval" />
        </change_format>
   </data>
  </outputs>
  <tests>
    <test>
      <param name="input" value="pileup_parser.6col.pileup"/>
      <output name="out_file1" file="pileup_parser.6col.20-3-yes-yes.pileup.out"/>
      <param name="type_select" value="six"/>
      <param name="qv_cutoff" value="20" />
      <param name="cvrg_cutoff" value="3" />
      <param name="snps_only" value="Yes"/>
      <param name="interval" value="Yes" />
      <param name="diff" value="No" />
      <param name="qc_base" value="Yes" />
    </test>
    <test>
      <param name="input" value="pileup_parser.6col.pileup"/>
      <output name="out_file1" file="pileup_parser.6col.20-3-yes-no.pileup.out"/>
      <param name="type_select" value="six"/>
      <param name="qv_cutoff" value="20" />
      <param name="cvrg_cutoff" value="3" />
      <param name="snps_only" value="Yes"/>
      <param name="interval" value="No" />
       <param name="diff" value="No" />
      <param name="qc_base" value="Yes" />
    </test>
    <test>
      <param name="input" value="pileup_parser.6col.pileup"/>
      <output name="out_file1" file="pileup_parser.6col.20-3-no-no.pileup.out"/>
      <param name="type_select" value="six"/>
      <param name="qv_cutoff" value="20" />
      <param name="cvrg_cutoff" value="3" />
      <param name="snps_only" value="No"/>
      <param name="interval" value="No" />
       <param name="diff" value="No" />
      <param name="qc_base" value="Yes" />
    </test>
    <test>
      <param name="input" value="pileup_parser.10col.pileup"/>
      <output name="out_file1" file="pileup_parser.10col.20-3-yes-yes.pileup.out"/>
      <param name="type_select" value="ten"/>
      <param name="qv_cutoff" value="20" />
      <param name="cvrg_cutoff" value="3" />
      <param name="snps_only" value="Yes"/>
      <param name="interval" value="Yes" />
       <param name="diff" value="No" />
      <param name="qc_base" value="Yes" />
    </test>
    <test>
      <param name="input" value="pileup_parser.10col.pileup"/>
      <output name="out_file1" file="pileup_parser.10col.20-3-yes-yes.pileup.out"/>
      <param name="type_select" value="manual"/>
      <param name="ref_base_column" value="3"/>
      <param name="read_bases_column" value="9"/>
      <param name="read_qv_column" value="10"/>
      <param name="cvrg_column" value="8"/>
      <param name="coord_column" value="2"/>
      <param name="qv_cutoff" value="20" />
      <param name="cvrg_cutoff" value="3" />
      <param name="snps_only" value="Yes"/>
      <param name="interval" value="Yes" />
      <param name="diff" value="No" />
      <param name="qc_base" value="Yes" />
    </test>
    <test>
      <param name="input" value="pileup_parser.10col.pileup"/>
      <output name="out_file1" file="pileup_parser.10col.20-3-yes-yes-yes-yes.pileup.out"/>
      <param name="type_select" value="manual"/>
      <param name="ref_base_column" value="3"/>
      <param name="read_bases_column" value="9"/>
      <param name="read_qv_column" value="10"/>
      <param name="cvrg_column" value="8"/>
      <param name="coord_column" value="2"/>
      <param name="qv_cutoff" value="20" />
      <param name="cvrg_cutoff" value="3" />
      <param name="snps_only" value="Yes"/>
      <param name="interval" value="Yes" />
      <param name="diff" value="Yes" />
      <param name="qc_base" value="Yes" />
    </test>
    <test>
      <param name="input" value="pileup_parser.10col.pileup"/>
      <output name="out_file1" file="pileup_parser.10col.20-3-yes-yes-yes-no.pileup.out"/>
      <param name="type_select" value="manual"/>
      <param name="ref_base_column" value="3"/>
      <param name="read_bases_column" value="9"/>
      <param name="read_qv_column" value="10"/>
      <param name="cvrg_column" value="8"/>
      <param name="coord_column" value="2"/>
      <param name="qv_cutoff" value="20" />
      <param name="cvrg_cutoff" value="3" />
      <param name="snps_only" value="Yes"/>
      <param name="interval" value="Yes" />
      <param name="diff" value="Yes" />
      <param name="qc_base" value="No" />
    </test>
</tests>
<help>

**What it does**

Allows one to find sequence variants and/or sites covered by a specified number of reads with bases above a set quality threshold. The tool works on six and ten column pileup formats produced with *samtools pileup* command. However, it also allows you to specify columns in the input file manually. The tool assumes the following:

- the quality scores follow phred33 convention, where input qualities are ASCII characters equal to the Phred quality plus 33.
- the pileup dataset was produced by the *samtools pileup* command (although you can override this by setting column assignments manually).

--------

**Types of pileup datasets**

The descriptions of the following pileup formats are largely based on information that can be found on the SAMTools_ documentation page. The 6- and 10-column variants are described below.

.. _SAMTools: http://samtools.sourceforge.net/pileup.shtml

**Six column pileup**::

    1    2  3  4        5        6
 ---------------------------------
 chrM  412  A  2       .,       II
 chrM  413  G  4     ..t,     IIIH
 chrM  414  C  4     ..Ta     III2
 chrM  415  C  4     TTTt     III7

where::

  Column Definition
 ------- ----------------------------
       1 Chromosome
       2 Position (1-based)
       3 Reference base at that position
       4 Coverage (# reads aligning over that position)
       5 Bases within reads
       6 Quality values (phred33 scale, see Galaxy wiki for more)

**Ten column pileup**

The `ten-column`__ pileup incorporates additional consensus information generated with the *-c* option of the *samtools pileup* command::


    1    2  3  4   5   6   7   8       9       10
 ------------------------------------------------
 chrM  412  A  A  75   0  25  2       .,       II
 chrM  413  G  G  72   0  25  4     ..t,     IIIH
 chrM  414  C  C  75   0  25  4     ..Ta     III2
 chrM  415  C  T  75  75  25  4     TTTt     III7

where::

  Column Definition
 ------- ----------------------------
       1 Chromosome
       2 Position (1-based)
       3 Reference base at that position
       4 Consensus bases
       5 Consensus quality
       6 SNP quality
       7 Maximum mapping quality
       8 Coverage (# reads aligning over that position)
       9 Bases within reads
      10 Quality values (phred33 scale, see Galaxy wiki for more)


.. __: http://samtools.sourceforge.net/cns0.shtml

------

**The output format**

The tool modifies the input dataset in two ways:

1. It appends five columns to the end of every reported line:

- Number of **A** variants
- Number of **C** variants
- Number of **G** variants
- Number of **T** variants
- Number of read bases covering this position, where quality is equal to or higher than the value set by **Do not consider read bases with  quality lower than** option.

Optionally, if **Print total number of differences?** is set to **Yes**, the tool will append the sixth column with the total number of deviants (see below).

2. If **Convert coordinates to intervals?** is set to **Yes**, the tool replaces the position column (typically the second column) with a pair of tab-delimited start/end values.

For example, if you are calling variants with base quality above 20 on this dataset::

 chrM  412  A  2       .,       II
 chrM  413  G  4     ..t,     III2
 chrM  414  C  4     ..Ta     III2
 chrM  415  C  4     TTTt     III7

you will get::

 chrM  413  G  4  ..t,  IIIH  0  0  2  1  3
 chrM  414  C  4  ..Ta  III2  1  1  0  1  3
 chrM  415  C  4  TTTt  III7  0  0  0  4  4

where::

  Column Definition
 ------- ----------------------------
       1 Chromosome
       2 Position (1-based)
       3 Reference base at that position
       4 Coverage (# reads aligning over that position)
       5 Bases within reads where
       6 Quality values (phred33 scale, see Galaxy wiki for more)
       7 Number of A variants
       8 Number of C variants
       9 Number of G variants
      10 Number of T variants
      11 Quality adjusted coverage:
      12 Number of read bases (i.e., # of reads) with quality above the set threshold
      13 Total number of deviants (if Convert coordinates to intervals? is set to yes)

if **Print total number of differences?** is set to **Yes**, you will get::

 chrM  413  G  4  ..t,  IIIH  0  0  2  1  3  1
 chrM  414  C  4  ..Ta  III2  1  2  0  1  3  2
 chrM  415  C  4  TTTt  III7  0  0  0  4  4  0

Note the additional column 13, that contains the number of deviant reads (e.g., there are two deviants, T and a, for position 414).

Finally, if **Convert coordinates to intervals?** is set to **Yes**, you will get one additional column with the end coordinate::

 chrM  412 413  G  4  ..t,  III2  0  0  2  1  3
 chrM  414 415  C  4  ..Ta  III2  1  2  0  1  3
 chrM  414 415  C  4  TTTt  III7  0  0  0  4  4

where::

  Column Definition
 ------- ----------------------------
       1 Chromosome
       2 Start position (0-based)
       3 End position (1-based)
       4 Reference base at that position
       5 Coverage (# reads aligning over that position)
       6 Bases within reads
       7 Quality values (phred33 scale, see Galaxy wiki for more)
       8 Number of A variants
       9 Number of C variants
      10 Number of G variants
      11 Number of T variants
      12 Quality adjusted coverage
      13 Total number of deviants (if Convert coordinates to intervals? is set to yes)


Note that in this case the coordinates of SNPs were converted to intervals, where the start coordinate is 0-based and the end coordinate in 1-based using the UCSC Table Browser convention.

Although three positions have variants in the original file (413, 414, and 415), only 413 and 415 are reported because the quality values associated with these two SNPs are above the threshold of 20. In the case of 414 the **a** allele has a quality value of 17 ( ord("2")-33 ), and is therefore not reported. Note that five columns have been added to each of the reported lines::

  chrM  413  G  4  ..t,  IIIH  0  0  2  1  3

Here, there is one variant, **t**. Because the fourth column represents **T** counts, it is incremented by 1. The last column shows that at this position, three reads have bases above the quality threshold of 20.

-----

**Example 1**: Just variants

In this mode, the tool only outputs the lines from the input datasets where at least one read contains a sequence variant with quality above the threshold set by the **Do not consider read bases with quality lower than** option. For example, suppose one has a pileup dataset like the following::

 chrM  412  A  2       .,       II
 chrM  413  G  4     ..t,     III2
 chrM  414  C  4     ..Ta     III2
 chrM  415  C  4     TTTt     III7

To call all variants (with no restriction by coverage) with quality above phred value of 20, we will need to set the parameters as follows:

.. image:: pileup_parser_help1.png

Running the tool with these parameters will return::

 chrM  413  G  4  ..t,  IIIH  0  0  0  1  3
 chrM  414  C  4  ..Ta  III2  0  2  0  1  3
 chrM  415  C  4  TTTt  III7  0  0  0  4  4

**Note** that position 414 is not reported because the *a* variant has associated quality value of 17 (because ord('2')-33 = 17) and is below the phred threshold of 20 set by the **Count variants with quality above this value** parameter.

-----

**Example 2**: Report everything

In addition to calling variants, it is often useful to know the quality adjusted coverage. Running the tool with these parameters:

.. image:: pileup_parser_help2.png

will report everything from the original file::

 chrM  412  A  2  .,    II    2  0  0  0  2
 chrM  413  G  4  ..t,  III2  0  0  2  1  3
 chrM  414  C  4  ..Ta  III2  0  2  0  1  3
 chrM  415  C  4  TTTt  III7  0  0  0  4  4

Here, you can see that although the total coverage at position 414 is 4 (column 4), the quality adjusted coverage is 3 (last column). This is because only three out of four reads have bases with quality above the set threshold of 20 (the actual qualities are III2 or, after conversion,  40, 40, 40, 17).

One can use the last column of this dataset to filter out (using Galaxy's **Filter** tool) positions where quality adjusted coverage (last column) is below a set threshold.

------

**Example 3**: Report everything and print total number of differences

If you set the **Print total number of differences?** to **Yes** the tool will print an additional column with the total number of reads where a devinat base is above the quality threshold. So, seetiing parametrs like this:

.. image:: pileup_parser_help3.png

will produce this::

 chrM  412  A  2  .,    II    2  0  0  0  2  0
 chrM  413  G  4  ..t,  III2  0  0  2  1  3  1
 chrM  414  C  4  ..Ta  III2  0  2  0  1  3  1
 chrM  415  C  4  TTTt  III7  0  0  0  4  4  0

-----

**Example 4**: Report everything, print total number of differences, and ignore qualities and read bases

Setting **Print quality and base string?** to **Yes** as shown here:

.. image:: pileup_parser_help4.png

will produce this::

 chrM  412  A  2  2  0  0  0  2  0
 chrM  413  G  4  0  0  2  1  3  1
 chrM  414  C  4  0  2  0  1  3  1
 chrM  415  C  4  0  0  0  4  4  0
</help>
<citations>
</citations>
</tool>