comparison umi-tools_extract.xml @ 15:27ac32a22ad2 draft

"planemo upload for repository https://github.com/galaxyproject/tools-iuc/tree/master/tools/umi_tools commit bf6a3aa532e8f9d122da4c1e39f3e256ae587b79"

14:9fa7803d1c51

<tool id="umi_tools_extract" name="UMI-tools extract" version="@VERSION@.2">
    <description>Extract UMI from fastq files</description>
    <macros>
        <import>macros.xml</import>
        <macro name="out_conditional">
            <actions>
                <conditional name="input_type.type">
                    <when value="paired_collection" >
                        <action type="format">
                            <option type="from_param" name="input_type.input_readpair" param_attribute="forward.ext" />
                        </action>
                    </when>
                    <when value="paired" >
                        <action type="format">
                            <option type="from_param" name="input_type.input_read1"  param_attribute="ext" />
                        </action>
                    </when>
                </conditional>
            </actions>
        </macro>
    </macros>
    <expand macro="requirements" />
    <command detect_errors="exit_code"><![CDATA[
    @COMMAND_LINK@

    umi_tools extract
            --extract-method='$extract_method.value'
            --bc-pattern='$bc_pattern'

            #if $input_type.type == 'single':
                #if $gz:
                    --stdin=input_single.gz
                    --stdout out.gz
                #else
                    --stdin=input_single.txt
                    --stdout '$out'
                #end if
            #else:
                #if $gz:
                    --stdin=input_read1.gz
                    --read2-in=input_read2.gz
                    --stdout out1.gz
                    --read2-out=out2.gz
                #else:
                    --stdin=input_read1.txt
                    --read2-in=input_read2.txt
                    --stdout '$out1'
                    --read2-out='$out2'
                #end if
                #if $input_type.barcode.barcode_select == "both_reads":
                    --split-barcode
                    --bc-pattern2='$input_type.barcode.bc_pattern2'
                #end if
            #end if

            #if $barcodes.use_barcodes.value == 'yes':
                --filter-cell-barcode
                --whitelist='$barcodes.filter_barcode_file'
                '$barcodes.filter_correct.value'
            #end if

            #if not $prime3:
                --3prime
            #end if
            #if $quality.quality_selector =='true':
                --quality-filter-threshold '$quality.quality_filter_threshold'
                --quality-encoding '$quality.quality_encoding'
            #end if
            #if $print_log == "1":
                --log='$out_log'
            #end if
        #if $gz:
            #if $input_type.type == 'single':
                && mv out.gz '$out'
            #else
                && mv out1.gz '$out1'
                && mv out2.gz '$out2'
            #end if
        #end if
    ]]></command>
    <inputs>
        <expand macro="input_types" />

        <conditional name="barcodes" >
            <param name="use_barcodes" argument="--filter-cell-barcode" type="select" label="Use Known Barcodes?" >
                <option value="yes">Yes</option>
                <option value="no" selected="true" >No</option>
            </param>
            <when value="no" />
            <when value="yes" >
                <param name="filter_barcode_file" type="data" format="tabular,tsv" label="Barcode File" />
                <param name="filter_correct" argument="--error-correct-cell" type="boolean" truevalue="--error-correct-cell" falsevalue="" checked="false" label="Apply correction to cell barcodes?" help="This only applies if your barcode file has two columns output from the umi_tools whitelist command." />
            </when>
        </conditional>

        <param name="extract_method" type="select" label="Method to extract barcodes" >
            <option value="regex">Regular Expressions</option>
            <option value="string" selected="true">String</option>
        </param>
            
        <param name="bc_pattern" argument="--bc-pattern" type="text" label="Barcode pattern for first read"
            help="Use this option to specify the format of the UMI/barcode. Use Ns to
                    represent the random positions and Xs to indicate the bc positions.
                    Bases with Ns will be extracted and added to the read name. Remaining
                    bases, marked with an X will be reattached to the read.">
            <expand macro="barcode_sanitizer" />
        </param>

        <param name="prime3" argument="--3prime" type="boolean" label="Is the barcode at the 5' end?"
            truevalue="1" falsevalue="0" checked="true"
            help="By default the barcode is assumed to be on the 5' end of the read, but
                use this option to sepecify that it is on the 3' end instead." />
        <param name="print_log" argument="-L" type="boolean" label="Output log?"
            truevalue="1" falsevalue="0" checked="true"
            help="Choose if you want to generate a text file containing logging information." />
        <conditional name="quality">
            <param name="quality_selector" type="select" label="Enable quality filter?" >
                <option value="false">No</option>
                <option value="true">Yes</option>
            </param>
            <when value="false">
            </when>
            <when value="true">
                <param name="quality_filter_threshold" label="Phred score threshold"
                    type="integer" value="20" argument="--quality-filter-threshold"
                    help="Remove reads where any UMI base quality score falls below this threshold." />
                <param name="quality_encoding" argument="--quality-encoding" type="select" label="Library type"
                    help="Quality score encoding. Choose from phred33 [33-77], phred64 [64-106] or solexa [59-106].">
                    <option value="phred33">phred33 [33-77]</option>
                    <option value="phred64">phred64 [64-106]</option>
                    <option value="solexa">solexa [59-106]</option>
                </param>
            </when>
        </conditional>
    </inputs>
    <outputs>
        <data name="out" format_source="input_single" label="Reads: ${tool.name} on ${on_string}" >
            <filter>input_type['type'] == "single"</filter>
        </data>
        <data name="out1" format_source="input_read1" label="Reads1: ${tool.name} on ${on_string}" >
            <filter>input_type['type'] != "single"</filter>
            <expand macro="out_conditional" />
        </data>
        <data name="out2" format_source="input_read2" label="Reads2: ${tool.name} on ${on_string}" >
            <filter>input_type['type'] != "single"</filter>
            <expand macro="out_conditional" />
        </data>
        
        <data name="out_log" format="txt">
            <filter>print_log == True</filter>
        </data>
    </outputs>
    <tests>
        <test expect_num_outputs="2">
            <param name="type" value="single" />
            <param name="input_single" value="t_R1.fastq" ftype="fastq" />
            <param name="bc_pattern" value="XXXNNN" />
            <param name="prime3" value="0" />
            <param name="quality_selector" value="true" />
            <param name="quality_filter_threshold" value="10" />
            <param name="quality_encoding" value="phred33" />
            <output name="out" file="out_SE.fastq" ftype="fastq" />
            <output name="out_log" >
                <assert_contents>
                    <has_text text="Input Reads: 100" />
                    <has_text text="umi quality: 28" />
                    <has_text text="Reads output: 72" />
                </assert_contents>
            </output>
        </test>
        <test expect_num_outputs="3">
            <param name="type" value="paired" />
            <param name="input_read1" value="t_R1.fastq.gz" ftype="fastq.gz" />
            <param name="input_read2" value="t_R2.fastq.gz" ftype="fastq.gz" />
            <param name="bc_pattern" value="NNNXXX" />
            <output name="out1" file="out_R1.fastq.gz" decompress="true" lines_diff="2" ftype="fastq.gz" />
            <output name="out2" file="out_R2.fastq.gz" decompress="true" lines_diff="2" ftype="fastq.gz" />
            <output name="out_log" >
                <assert_contents>
                    <has_text text="Input Reads: 100" />
                    <has_text text="Reads output: 100" />
                </assert_contents>
            </output>
        </test>
        <test expect_num_outputs="3">
            <param name="type" value="paired_collection" /> <!-- same as before, but uncompressed -->
            <param name="paired_type" value="no" />
            <param name="input_readpair" >
                <collection type="paired" >
                    <element name="forward" ftype="fastq" value="t_R1.fastq" />
                    <element name="reverse" ftype="fastq" value="t_R2.fastq" />
                </collection>
            </param>
            <param name="bc_pattern" value="NNNXXX" />
            <output name="out1" file="out_R1.fastq" ftype="fastq" />
            <output name="out2" file="out_R2.fastq" ftype="fastq" />
            <output name="out_log" >
                <assert_contents>
                    <has_text text="Input Reads: 100" />
                    <has_text text="Reads output: 100" />
                </assert_contents>
            </output>
        </test>
        <test expect_num_outputs="3">
            <param name="type" value="paired" />
            <param name="input_read1" value="scrb_seq_fastq.1.gz" ftype="fastq.gz" />
            <param name="input_read2" value="scrb_seq_fastq.2.gz" ftype="fastq.gz" />
            <param name="extract_method" value="string" />
            <param name="bc_pattern" value="CCCCCCNNNNNNNNNN" />
            <param name="use_barcodes" value="yes" />
            <param name="filter_barcode_file" value="scrb_seq_barcodes" />
            <output name="out2" file="scrb_extract.fastq.gz" decompress="true"  ftype="fastq.gz" />
        </test>
        <test expect_num_outputs="3"><!-- same as above but with regex barcode-->
            <param name="type" value="paired" />
            <param name="input_read1" value="scrb_seq_fastq.1.gz" ftype="fastq.gz" />
            <param name="input_read2" value="scrb_seq_fastq.2.gz" ftype="fastq.gz" />
            <param name="extract_method" value="regex" />
            <param name="bc_pattern" value="^(?P&lt;cell_1&gt;.{6})(?P&lt;umi_1&gt;.{10})" />
            <param name="use_barcodes" value="yes" />
            <param name="filter_barcode_file" value="scrb_seq_barcodes" />
            <output name="out2" file="scrb_extract.fastq.gz" decompress="true" ftype="fastq.gz" />
        </test>
        <test expect_num_outputs="2"><!-- CelSeq2 example -->
            <param name="type" value="paired" />
            <param name="input_read1" value="read_R1.200.gz" ftype="fastq.gz" />
            <param name="input_read2" value="read_R2.200.gz" ftype="fastq.gz" />
            <param name="extract_method" value="string" />
            <param name="bc_pattern" value="NNNNNNCCCCCC" />
            <output name="out1" file="read_R1.200_extracted.fastq.gz" ftype="fastq.gz" decompress="true" lines_diff="1" />
            <output name="out2" file="read_R2.200_extracted.fastq.gz" ftype="fastq.gz" decompress="true" lines_diff="1" />
            <param name="print_log" value="false"/>
        </test>
    </tests>
    <help><![CDATA[


UMI-tools extract.py - Extract UMI from fastq
=============================================

Purpose
-------

Extract UMI barcode from a read and add it to the read name, leaving
any sample barcode in place. Can deal with paired end reads and UMIs
split across the paired ends

Options
-------

--split-barcode
       By default the UMI is assumed to be on the first read. Use this
       option if the UMI is contained on both reads and specify the
       pattern of the barcode/UMI on the second read using the option
       ``--bc-pattern2``

--bc-pattern
       Use this option to specify the format of the UMI/barcode. Use Ns to
       represent the random positions and Xs to indicate the bc positions.
       Bases with Ns will be extracted and added to the read name. Remaining
       bases, marked with an X will be reattached to the read.

       E.g. If the pattern is NNXXNN,
       Then the read:

       @HISEQ:87:00000000 read1
       AAGGTTGCTGATTGGATGGGCTAG
       DA1AEBFGGCG01DFH00B1FF0B
       +

       will become:
       @HISEQ:87:00000000_AATT read1
       GGGCTGATTGGATGGGCTAG
       1AFGGCG01DFH00B1FF0B
       +

--bc-pattern2
       Use this option to specify the format of the UMI/barcode for
       the second read pair if required. If --bc-pattern2 is not
       supplied, this defaults to the same pattern as --bc-pattern

--3prime
       By default the barcode is assumed to be on the 5' end of the read, but
       use this option to sepecify that it is on the 3' end instead

-L
       Specify a log file to retain logging information and final statistics

--split-barcode
       barcode is split across read pair

--quality-filter-threshold=QUALITY_FILTER_THRESHOLD
       Remove reads where any UMI base quality score falls
       below this threshold
--quality-encoding=QUALITY_ENCODING
       Quality score encoding. Choose from phred33[33-77]
       phred64 [64-106] or solexa [59-106]

Usage:
------

For single ended reads:
        umi_tools extract --bc-pattern=[PATTERN] -L extract.log [OPTIONS]

reads from stdin and outputs to stdout.

For paired end reads:
        umi_tools extract --bc-pattern=[PATTERN] --read2-in=[FASTQIN] --read2-out=[FASTQOUT] -L extract.log [OPTIONS]

reads end one from stdin and end two from FASTQIN and outputs end one to stdin
and end two to FASTQOUT.

    ]]></help>
    <expand macro="citations" />
</tool>

15:27ac32a22ad2

<tool id="umi_tools_extract" name="UMI-tools extract" version="@TOOL_VERSION@+galaxy@VERSION_SUFFIX@" profile="@PROFILE@">
    <description>Extract UMI from fastq files</description>
    <expand macro="bio_tools"/>
    <macros>
        <import>macros.xml</import>
    </macros>
    <expand macro="requirements" />
    <command detect_errors="exit_code"><![CDATA[
    @COMMAND_LINK@

    umi_tools extract
        
        @FASTQ_BARCODE_EXTRACTION_OPTIONS@
        #if $input_type_cond.input_type == 'single':
            #if $gz:
                --stdin=input_single.gz
                --stdout out.gz
            #else
                --stdin=input_single.txt
                --stdout '$out'
            #end if
        #else:
            #if $gz:
                --stdin=input_read1.gz
                --read2-in=input_read2.gz
                --stdout out1.gz
                --read2-out=out2.gz
            #else:
                --stdin=input_read1.txt
                --read2-in=input_read2.txt
                #if $input_type_cond.input_type == 'paired'
                    --stdout '$out'
                    --read2-out='$out2'
                #else
                    --stdout '$out_paired_collection.forward'
                    --read2-out='$out_paired_collection.reverse'
                #end if
            #end if
            $input_type_cond.reconcile_pairs
        #end if

        #if $whitelist
            --whitelist='$whitelist'
        #end if
        #if $blacklist
            --blacklist='$blacklist'
        #end if
        $error_correct_cell.value

        #if $quality.quality_selector =='true':
            #if str($quality.quality_filter_threshold) != ''
                --quality-filter-threshold '$quality.quality_filter_threshold'
            #end if
            #if str($quality.quality_filter_mask) != ''
                --quality-filter-mask '$quality.quality_filter_mask'
            #end if
            #if $input_type_cond.input_type != 'paired_collection'
                #set input=$input_type_cond.input_read1
            #else
                #set input=$input_type_cond.input_readpair.forward
            #end if
            --quality-encoding
            #if $input.ext.startswith("fastqillumina")
                phred64
            #else if $input.ext.startswith("fastqsolexa")
                solexa
            #else
                phred33
            #end if
        #end if
        @LOG@
        #if $gz:
            #if $input_type_cond.input_type == 'single':
                && mv out.gz '$out'
            #else if $input_type_cond.input_type == 'paired' 
                && mv out1.gz '$out'
                && mv out2.gz '$out2'
            #else
                && mv out1.gz '$out_paired_collection.forward'
                && mv out2.gz '$out_paired_collection.reverse'
            #end if
        #end if
    ]]></command>
    <inputs>
        <expand macro="input_types">
                <param argument="--reconcile-pairs" type="boolean" truevalue="--reconcile-pairs" falsevalue="" checked="false" label="Allow unpaired reads" help="Allow the presences of reads in read2 input that are not present in read1 input. This allows cell barcode filtering of read1s without considering read2s" />
        </expand>
        <expand macro="fastq_barcode_extraction_options_macro"/>

        <param argument="--whitelist" type="data" optional="true" format="tabular,tsv" label="Allowlist of accepted barcodes" />
        <param argument="--blacklist" type="data" optional="true" format="tabular,tsv" label="Denylist of accepted barcodes" />
        <param argument="--error-correct-cell" type="boolean" truevalue="--error-correct-cell" falsevalue="" checked="false" label="Apply correction to cell barcodes?" help="This only applies if your barcode file has two columns output from the umi_tools whitelist command" />

        <conditional name="quality">
            <param name="quality_selector" type="select" label="Enable quality filter?" >
                <option value="false">No</option>
                <option value="true">Yes</option>
            </param>
            <when value="false">
            </when>
            <when value="true">
                <param argument="--quality-filter-threshold" label="Phred score threshold"
                    type="integer" value="" optional="true"
                    help="Remove reads where any UMI base quality score falls below this threshold" />
                <param argument="--quality-filter-mask" label="Mask UMI bases below threshold"
                    type="integer" value="" optional="true"
                    help="If a UMI base has a quality below this threshold,
                    replace the base with 'N'" />
            </when>
        </conditional>
        <expand macro="log_input_macro"/>
    </inputs>
    <outputs>
        <data name="out" format_source="input_read1" label="${tool.name} on ${on_string}: Reads" >
            <filter>input_type_cond['input_type'] in ['single', 'paired']</filter>
        </data>
        <data name="out2" format_source="input_read2" label="${tool.name} on ${on_string}: Reads2" >
            <filter>input_type_cond['input_type'] == 'paired'</filter>
        </data>
        <collection name="out_paired_collection" type="paired" label="${tool.name} on ${on_string}: Reads">
            <data name="forward" format_source="input_readpair" />
            <data name="reverse" format_source="input_readpair" />
            <filter>input_type_cond['input_type'] == 'paired_collection'</filter>
        </collection>
        <expand macro="log_output_macro"/>
    </outputs>
    <tests>
        <test expect_num_outputs="2">
            <conditional name="input_type_cond">
                <param name="input_type" value="single" />
                <param name="input_read1" value="t_R1.fastq" ftype="fastqsanger" />
                <param name="bc_pattern" value="XXXNNN" />
            </conditional>
            <conditional name="extract_method_cond">
                <param name="prime3" value="true" />
            </conditional>
            <param name="quality_selector" value="true" />
            <param name="quality_filter_threshold" value="10" />
            <param name="quality_encoding" value="phred33" />
            <param name="log" value="true"/>
            <output name="out" file="out_SE.fastq" ftype="fastqsanger" />
            <output name="out_log" >
                <assert_contents>
                    <has_text text="Input Reads: 100" />
                    <has_text text="umi quality: 28" />
                    <has_text text="Reads output: 72" />
                </assert_contents>
            </output>
        </test>
        <test expect_num_outputs="3">
            <conditional name="input_type_cond">
                <param name="input_type" value="paired" />
                <param name="input_read1" value="t_R1.fastq.gz" ftype="fastqsanger.gz" />
                <param name="input_read2" value="t_R2.fastq.gz" ftype="fastqsanger.gz" />
                <param name="bc_pattern" value="NNNXXX" />
            </conditional>
            <param name="log" value="true"/>
            <output name="out" file="out_R1.fastq.gz" decompress="true" lines_diff="2" ftype="fastqsanger.gz" />
            <output name="out2" file="out_R2.fastq.gz" decompress="true" lines_diff="2" ftype="fastqsanger.gz" />
            <output name="out_log" >
                <assert_contents>
                    <has_text text="Input Reads: 100" />
                    <has_text text="Reads output: 100" />
                </assert_contents>
            </output>
        </test>
        <test expect_num_outputs="4">
            <conditional name="input_type_cond">
                <param name="input_type" value="paired_collection" /> <!-- same as before, but uncompressed -->
                <param name="paired_type" value="no" />
                <param name="input_readpair">
                    <collection type="paired" >
                        <element name="forward" ftype="fastqsanger" value="t_R1.fastq" />
                        <element name="reverse" ftype="fastqsanger" value="t_R2.fastq" />
                    </collection>
                </param>
                <param name="bc_pattern" value="NNNXXX" />
            </conditional>
            <param name="log" value="true"/>
            <output_collection name="out_paired_collection" type="paired">
                <element name="forward" file="out_R1.fastq" ftype="fastqsanger" />
                <element name="reverse" file="out_R2.fastq" ftype="fastqsanger" />
            </output_collection>
            <output name="out_log" >
                <assert_contents>
                    <has_text text="Input Reads: 100" />
                    <has_text text="Reads output: 100" />
                </assert_contents>
            </output>
        </test>
        <test expect_num_outputs="3">
            <conditional name="input_type_cond">
                <param name="input_type" value="paired" />
                <param name="input_read1" value="scrb_seq_fastq.1.gz" ftype="fastqsanger.gz" />
                <param name="input_read2" value="scrb_seq_fastq.2.gz" ftype="fastqsanger.gz" />
                <param name="bc_pattern" value="CCCCCCNNNNNNNNNN" />
            </conditional>
            <param name="extract_method" value="string" />
            <param name="whitelist" value="scrb_seq_barcodes" />
            <param name="log" value="true"/>
            <output name="out2" file="scrb_extract.fastq.gz" decompress="true"  ftype="fastqsanger.gz" />
        </test>
        <test expect_num_outputs="3"><!-- same as above but with regex barcode-->
            <conditional name="input_type_cond">
                <param name="input_type" value="paired" />
                <param name="input_read1" value="scrb_seq_fastq.1.gz" ftype="fastqsanger.gz" />
                <param name="input_read2" value="scrb_seq_fastq.2.gz" ftype="fastqsanger.gz" />
                <param name="bc_pattern" value="^(?P&lt;cell_1&gt;.{6})(?P&lt;umi_1&gt;.{10})" />
            </conditional>
            <param name="extract_method" value="regex" />
            <param name="whitelist" value="scrb_seq_barcodes" />
            <param name="log" value="true"/>
            <output name="out2" file="scrb_extract.fastq.gz" decompress="true" ftype="fastqsanger.gz" />
        </test>
        <test expect_num_outputs="2"><!-- CelSeq2 example -->
            <conditional name="input_type_cond">
                <param name="input_type" value="paired" />
                <param name="input_read1" value="read_R1.200.gz" ftype="fastqsanger.gz" />
                <param name="input_read2" value="read_R2.200.gz" ftype="fastqsanger.gz" />
                <param name="bc_pattern" value="NNNNNNCCCCCC" />
            </conditional>
            <param name="extract_method" value="string" />
            <output name="out" file="read_R1.200_extracted.fastq.gz" ftype="fastqsanger.gz" decompress="true" lines_diff="1" />
            <output name="out2" file="read_R2.200_extracted.fastq.gz" ftype="fastqsanger.gz" decompress="true" lines_diff="1" />
        </test>
    </tests>
    <help><![CDATA[

extract - Extract UMI from fastq
================================

Extract UMI barcode from a read and add it to the read name, leaving
any sample barcode in place

Can deal with paired end reads and UMIs
split across the paired ends. Can also optionally extract cell
barcodes and append these to the read name also. See the section below
for an explanation for how to encode the barcode pattern(s) to
specficy the position of the UMI +/- cell barcode.


Filtering and correcting cell barcodes
--------------------------------------

``umi_tools extract`` can optionally filter cell barcodes against a user-supplied
whitelist (``--whitelist``). If a whitelist is not available for your data,
e.g
if you have performed droplet-based scRNA-Seq, you can use the
whitelist tool.

Cell barcodes which do not match the whitelist (user-generated or
automatically generated) can also be optionally corrected using the
``--error-correct-cell`` option.

The whitelist should be in  the following format (tab-separated)::

        AAAAAA    AGAAAA
        AAAATC
        AAACAT
        AAACTA    AAACTN,GAACTA
        AAATAC
        AAATCA    GAATCA
        AAATGT    AAAGGT,CAATGT

Where column 1 is the whitelisted cell barcodes and column 2 is
the list (comma-separated) of other cell barcodes which should be
corrected to the barcode in column 1. If the ``--error-correct-cell``
option is not used, this column will be ignored. Any additional columns
in the whitelist input, such as the counts columns from the output of
umi_tools whitelist, will be ignored.

@FASTQ_BARCODE_EXTRACTION_HELP@

    ]]></help>
    <expand macro="citations" />
</tool>