<tool id="je_clip" name="Je-Clip" version="1.0">
    <description>clips Unique Molecular Identifiers (UMIs) from fastq files</description>
    <macros>
        <import>macros.xml</import>
    </macros>
    <stdio>
        <exit_code range="1:" level="fatal" description="Tool exception" />
    </stdio>
    <version_command>echo '1.0'</version_command>
    <command interpreter="bash">
<![CDATA[
    je clip

    ## Fastq inputs
    @single_or_paired_cmd@
    #if str( $library.type ) != "single":
        BPOS=${library.BPOS}
    #end if

    @common_options_cmd@
    @barcode_len_cmd@
    ADD=${ADD}
    #if str($ADD) == "false":
        BARCODE_RESULT_FILENAME=$BARCODE_RESULT_FILENAME
    #end if

    OF1=${OF1}
    #if str( $library.type ) != "single":
        OF2=${OF2}
    #end if

    FORCE=true
]]>
    </command>
    <inputs>
        <!-- single/paired -->
        <expand macro="single_or_paired_general">
            <param name="BPOS" type="select" label="Barcode read position (BPOS)" help="where are the barcodes.">
                <option value="READ_1" selected="true">READ_1 (beginning of read from the first fastq file)</option>
                <option value="READ_2">READ_2 (beginning of read from the second fastq file)</option>
                <option value="BOTH">BOTH (beginning of both reads)</option>
            </param>
        </expand>
        <expand macro="barcode_len_option"/>
        <param name="ADD" type="boolean"
            label="Add matched barcode at the end of the read header (ADD)"
            truevalue="true"
            falsevalue="false"
            checked="true"
        />

        <expand macro="common_options"/>


    </inputs>
    <outputs>
        <data name="BARCODE_RESULT_FILENAME" format="tabular" label="Je-Clipped Barcodes"/>
        <data name="OF1" format_source="input_1" label="Je-Clipped {on_string}"/>
        <data name="OF2" format_source="input_1" label="Je-Clipped {on_string}">
            <filter>(type != "single")</filter>
        </data>
    </outputs>

    <tests>
        <test>
            <!-- simple test on single end data -->
            <param name="type" value="single"/>
            <param name="input_1" value="file_1_sequence.txt" ftype="fastqsanger"/>
            <param name="LEN" value="6"/>
            <param name="ADD" value="false"/>
            <output name="BARCODE_RESULT_FILENAME" file="clip_barcode_result_file.txt"/>
            <output name="OF1" file="clip_dataset1_SE.fastq"/>
        </test>
        <test>
            <!-- more complex test on paired end data with different barcode for fwd/rev -->
            <param name="type" value="paired"/>
            <param name="input_1" value="file_1_sequence.txt" ftype="fastqsanger"/>
            <param name="input_2" value="file_2_sequence.txt" ftype="fastqsanger"/>
            <param name="LEN" value="6"/>
            <param name="BPOS" value="BOTH"/>
            <output name="OF1" file="clip_dataset1_PE.fastq"/>
            <output name="OF2" file="clip_dataset2_PE.fastq"/>
        </test>
    </tests>


  <help>
<![CDATA[
**What it does**

Je clip: Clips barcodes or Unique Molecular Identifiers (UMIs) from the input fastq files
Input files are fastq files, and can be in gzip compressed format.

Author: Charles Girardot  (charles.girardot@embl.de).

Wrapper by: Jelle Scholtalbers (jelle.scholtalbers@embl.de).

------

**Know what you are doing**

.. class:: warningmark

  You will want to read the `documentation`__.

  .. __: http://gbcs.embl.de/portal/Je

------

**Parameter list**

This is an exhaustive list of options::

  FASTQ_FILE1=File
  F1=File

    Input fastq file (optionally gzipped) for single end data, or first read in paired end data.
    Required.

  FASTQ_FILE2=File
  F2=File

    Input fastq file (optionally gzipped) for the second read of paired end data.
    Default value: null.

  BCLEN=String
  LEN=String

    Length of the barcode sequences. When BARCODE_READ_POS == BOTH, two distinct lengths can
    be provided using the syntax LEN=X:Z where X and Z are 2 integers representing the
    barcode length for read_1 and read_2 respectively.
    Required.

  BARCODE_READ_POS=BarcodePosition
  BPOS=BarcodePosition

    Reads containing the sequence (i.e. UMIs) to clip:
      READ_1 (beginning of read from FASTQ_FILE_1),
      READ_2 (beginning of read from FASTQ_FILE_2),
      BOTH (beginning of both reads).

    Automatically set to READ_1 in single end mode and BOTH in paired end mode. Actually not
    relevant for single end data
    Default value: BOTH. This option can be set to 'null' to clear the default value.
    Possible values: {READ_1, READ_2, BOTH, NONE}

  ADD_BARCODE_TO_HEADER=Boolean
  ADD=Boolean

    Should clipped UMIs be added to the read header (at the end); apply to both barcodes when
    BPOS=BOTH.
    If ADD=true, the string ':barcode' is added at the end of the read header with a ':'
    added only if current read header does not end with ':'.
    If both reads of the pair contains a UMI (i.e. BARCODE_READ_POS == BOTH), the UMI from
    the second read is also added to the read header.
    Else, the header of the read without UMI receives the UMI from the other read.
    For example:
      @D3FCO8P1:178:C1WLBACXX:7:1101:1836:1965 2:N:0:
    becomes
      @D3FCO8P1:178:C1WLBACXX:7:1101:1836:1965 2:N:0:BARCODE
    Default value: true. This option can be set to 'null' to clear the default value.
    Possible values: {true, false}

  ENSURE_IDENTICAL_HEADER_NAMES=Boolean
  SAME_HEADERS=Boolean

    Makes sure headers of both reads of a pair are identical.
    Read name (or headers) will follow the pattern (for both reads of a pair):
      @D3FCO8P1:178:C1WLBACXX:7:1101:1836:1965 CLIPPED_SEQ_FROMREAD1:CLIPPED_SEQ_FROMREAD2
    This option only makes sense in paired end mode and ADD=true.Some (if not all) mappers
    will indeed complain when read headers of a read pair are not identical.
    When SAME_HEADERS=FALSE and the RCHAR is used, read headers look like this:
      HISEQ:44:C6KC0ANXX:5:1101:1491:1994:1:N:0:TGGAGTAG
      HISEQ:44:C6KC0ANXX:5:1101:1491:1994:3:N:0:CGTTGTAT

    SAME_HEADERS=true will instead generates the following identical header for both reads :
      HISEQ:44:C6KC0ANXX:5:1101:1491:1994:TGGAGTAG:CGTTGTAT
    Note that we also clipped the useless '1:N:0' amd '3:N:0' as they also result in
    different headers
    Important : this option will force RCHAR=: UNLESS you specify RCHAR=null ; in which case
    a space will be preserved i.e.:
      HISEQ:44:C6KC0ANXX:5:1101:1491:1994 TAGAACAC:TGGAGTAG:CGTTGTAT

    Default value: true.
    This option can be set to 'null' to clear the default value. Possible values: {true,
    false}

  READ_NAME_REPLACE_CHAR=String
  RCHAR=String

    Replace spaces in read name/header using provided character.
    This is needed when you need to retain ADDed barcode in read name/header during mapping
    as everything after space in read name is usually clipped in BAM files.
    For example, with RCHAR=':':
      @D3FCO8P1:178:C1WLBACXX:7:1101:1836:1965 1:N:0:
    becomes
      @D3FCO8P1:178:C1WLBACXX:7:1101:1836:1965:1:N:0:BARCODE

    Default value: ':'. This option can be set to 'null' to clear the default value.

  XTRIMLEN=String
  XT=String

    Optional extra number of base(s) to be trimmed right after the barcode. These extra bases
    are not added to read headers.
    When running paired-end, two distinct values can be given using the syntax XT=X:Z where X
    and Z are 2 integers to use for read_1 and read_2 respectively. Note that even when
    BPOS=READ_1 or BPOS=READ_2, a X:Y synthax can be given to trim the read w/o barcode to
    end up with reads of identical length (note that this can also be operated using ZT). If
    a unique value is given, e.g. XT=1, while running paired-end the following rule applies :
      (1) BPOS=READ_1 or BPOS=READ_2, no trim is applied at the read w/o barcode
      (2) BPOS=BOTH, the value is used for both reads.
    Note that XT=null is like XT=0.
    Default value: 0. This option can be set to 'null' to clear the default value.

  ZTRIMLEN=String
  ZT=String

    Optional extra number of bases to be trimmed from the read end i.e. 3' end. These extra
    bases are not added to read headers.
    When running paired-end, two distinct values can be given here using the syntax ZT=X:Z
    where X and Z are 2 integers to use for read_1 and read_2 respectively. Note that even
    when BPOS=READ_1 or BPOS=READ_2, a X:Y synthax can be given to trim the read w/o barcode
    as to end up with reads of the same length (note that this can also be operated using
    XT). Note that if a single value is passed, the value always applies to both reads in
    paired-end mode without further consideration.

    Default value: 0. This option can be set to 'null' to clear the default value.

  BARCODE_RESULT_FILENAME=String
  BF=String

    Optional file name where to write clipped barcodes, default name is clipped_barcodes.txt.
    This file is automatically created if ADD=FALSE i.e. even if this option is not provided
    by user (and always created if this option is given).
    File format is tab delimited with:
    ``read header (col 1)   barcode from read_1 (col 2) barcode quality from read_1 (col 2)``
    + barcode + quality from read_2 (col 4 and 5 respectively) when relevant.
    Can either be a name (in which case the file will be created in the output dir) or a full path.
    Default value: null.

]]>
  </help>

</tool>