view fastqc_site.xml @ 12:68ea2ebbf866 draft

add boxplot for per base sequence quality

<tool id="fastqc_site" name="Fastqc Site" version="2.0.0">
    <requirements>
        <requirement type="package" version="1.15.0.6-0">pandoc</requirement>
        <requirement type="package" version="1.14.1">bioconductor-deseq2</requirement>
        <requirement type="package" version="1.20.0">r-getopt</requirement>
        <requirement type="package" version="1.3">r-rmarkdown</requirement>
        <requirement type="package" version="1.8.4">r-plyr</requirement>
        <requirement type="package" version="1.1.0">r-stringr</requirement>
        <requirement type="package" version="0.5.0">r-highcharter</requirement>
        <requirement type="package" version="0.2">r-dt</requirement>
        <requirement type="package" version="1.4.2">r-reshape2</requirement>
        <requirement type="package" version="4.5.6">r-plotly</requirement>
        <requirement type="package" version="0.2.0.1">r-formattable</requirement>
        <requirement type="package" version="0.3.5">r-htmltools</requirement>
        <requirement type="package" version="0.11.5">fastqc</requirement>
    </requirements>
    <stdio>
        <!--redirecting stderr to a file. "XXX" is used to match with nothing so that tool running won't be interrupted during testing-->
        <regex match="XXX"
               source="stderr"
               level="warning"
               description="Check the warnings_and_errors.txt file for more details."/>
    </stdio>
    <command>
        <![CDATA[

        Rscript '${__tool_directory__}/fastqc_site_render.R'

            ## 1. input data
            -e $echo
            -r $reads_1
            -n '$reads_1.name'
            -R $reads_2
            -N '$reads_2.name'
            -c $contaminants
            -l $limits

            ## 2. output report and report site directory
		    -o $report
		    -d $report.files_path
		    -s $sink_message

		    ## 3. Rmd templates sitting in the tool directory

		    ## _site.yml and index.Rmd template files
            -S '${__tool_directory__}/_site.yml'
            -I '${__tool_directory__}/index.Rmd'

            ## other Rmd body template files
		    -A '${__tool_directory__}/01_evaluation_overview.Rmd'
		    -B '${__tool_directory__}/02_per_base_sequence_quality.Rmd'
		    -C '${__tool_directory__}/03_per_tile_sequence_quality.Rmd'
		    -D '${__tool_directory__}/04_per_sequence_quality_score.Rmd'
		    -E '${__tool_directory__}/05_per_base_sequence_content.Rmd'
		    -F '${__tool_directory__}/06_per_sequence_gc_content.Rmd'
		    -G '${__tool_directory__}/07_per_base_n_content.Rmd'
		    -H '${__tool_directory__}/08_sequence_length_distribution.Rmd'
		    -J '${__tool_directory__}/09_sequence_duplication_levels.Rmd'
		    -K '${__tool_directory__}/10_adapter_content.Rmd'
		    -L '${__tool_directory__}/11_kmer_content.Rmd'


        ]]>
    </command>
    <inputs>
        <param format="fastq,fastq.gz,fastq.bz2,bam,sam" name="reads_1" type="data" optional="false"
               label="Short reads before trimming"
               help="Short reads data from history. This could be reads before trimming."/>
        <param format="fastq,fastq.gz,fastq.bz2,bam,sam" name="reads_2" type="data"
               label="Short reads after trimming"
               help="Short reads data from history. This could be reads after trimming."/>
        <param name="contaminants" type="data" format="tabular" optional="true" label="Contaminant list"
               help="Specifies a non-default file which contains the list of adapter sequences which will be explicitly
                     searched against the library. The file must contain sets of named adapters
                     in the form name[tab]sequence.  Lines prefixed with a hash will be ignored."/>
        <param name="limits" type="data" format="txt" optional="true" label="Submodule and Limit specifing file"
               help="Specifies a non-default file which contains a set of criteria
                    which will be used to determine the warn/error limits for the
                    various modules.  This file can also be used to selectively
                    remove some modules from the output all together.  The format
                    needs to mirror the default limits.txt file found in the
                    Configuration folder."/>
        <param type="boolean" name="echo" truevalue="TRUE" falsevalue="FALSE" checked="false"
               label="Display analysis code in report?"/>
    </inputs>
    <outputs>
        <data format="html" name="report" label="fastqc site"/>
        <data format="txt" name="sink_message" label="Warnings and Errors" from_work_dir="warnings_and_errors.txt"/>
    </outputs>
    <citations>
        <citation type="bibtex">
            @misc{bioinformatics2014fastqc,
            title={FastQC},
            author={Bioinformatics, Babraham},
            year={2014}
            }
        </citation>
        <citation type="bibtex">
            @article{allaire2016rmarkdown,
            title={rmarkdown: Dynamic Documents for R, 2016},
            author={Allaire, J and Cheng, Joe and Xie, Yihui and McPherson, Jonathan and Chang, Winston and Allen, Jeff
            and Wickham, Hadley and Atkins, Aron and Hyndman, Rob},
            journal={R package version 0.9},
            volume={6},
            year={2016}
            }
        </citation>
        <citation type="bibtex">
            @book{xie2015dynamic,
            title={Dynamic Documents with R and knitr},
            author={Xie, Yihui},
            volume={29},
            year={2015},
            publisher={CRC Press}
            }
        </citation>
        <citation type="bibtex">
            @misc{plotly2017,
            title = {plotly: Create Interactive Web Graphics via 'plotly.js'},
            author = {Carson Sievert and Chris Parmer and Toby Hocking and Scott Chamberlain and Karthik Ram and
            Marianne Corvellec and Pedro Despouy},
            year = {2017},
            note = {R package version 4.6.0},
            url = {https://CRAN.R-project.org/package=plotly},
            }
        </citation>
        <citation type="bibtex">
            @article{ewels2016multiqc,
            title={MultiQC: summarize analysis results for multiple tools and samples in a single report},
            author={Ewels, Philip and Magnusson, M{\aa}ns and Lundin, Sverker and K{\"a}ller, Max},
            journal={Bioinformatics},
            volume={32},
            number={19},
            pages={3047--3048},
            year={2016},
            publisher={Oxford University Press}
            }
        </citation>
    </citations>
</tool>