view nanopolish_polya.xml @ 5:4857d5bcce52 draft default tip

planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/tools/nanopolish commit de2370d1a385731b3c65f1dcc44e7b8558da8fd4

<tool id="nanopolish_polya" name="Nanopolish polyA" version="@VERSION@+galaxy@VERSION_SUFFIX@" profile="@PROFILE@">
    <description>- Estimate the length of the poly-A tail on direct RNA reads.</description>
    <macros>
        <import>macros.xml</import>
    </macros>
    <expand macro="requirements" />
    <command detect_errors="exit_code"><![CDATA[
        @PREPROCESS_INPUTS@
        
        nanopolish polya
            -r reads.fasta
            -b reads.bam
            -g genome.fa
            --threads "\${GALAXY_SLOTS:-4}"
            #if $w and str($w).strip():
              -w "${w}"
            #end if  
        > polya_results.tsv
    ]]></command>
    <inputs>
      <!-- index inputs -->
        <param type="data" name="input_merged" format="fasta,fastq" label="Basecalled merged reads.fa"/>
        <param type="data" name="input_reads_raw" format="fast5.tar.gz,fast5.tar.bz2,fast5.tar" label="Flat archive file of raw fast5 files"/>

        <param type="data" argument="-b" format="bam" label="Reads aligned to the reference genome" />
        <conditional name="reference_source">
          <param name="reference_source_selector" type="select" label="Load reference genome from">
            <option value="cached">Local cache</option>
            <option value="history">History</option>
          </param>
          <when value="cached">
            <param name="ref_file" type="select" label="Using reference genome" help="REFERENCE_SEQUENCE">
              <options from_data_table="all_fasta">
              </options>
              <validator type="no_options" message="A built-in reference genome is not available for the build associated with the selected input file"/>
            </param>
          </when>
          <when value="history">
            <param name="ref_file" type="data" format="fasta" label="Use the following dataset as the reference sequence" help="REFERENCE_SEQUENCE; You can upload a FASTA sequence to the history and use it as reference" />
          </when>
        </conditional>

        <section name="adv" title="Optional data inputs">
            <!-- optional inputs -->
            <param type="data" name="input_seq_summary" format="txt" optional="true" label="Sequencing summary file from albacore" help="(-s)"/>       
        </section>
        
        <param argument="-w" type="text" optional="true"
            label="only compute the poly-A lengths for reads in window STR (format: ctg:start_id-end_id)" />

    </inputs>

    <outputs>
        <data name="polya_results" format="tabular" from_work_dir="polya_results.tsv" label="called methylation sites" />
    </outputs>
    <tests>
        <test>
            <param name="input_merged" ftype="fastq" value="30xpolyA-small-subset.fastq" />
            <param name="input_reads_raw" ftype="fast5.tar.gz" value="fast5_files_30xpolyA-small-subset.tar.gz" />
            <param name="b" value="30xpolyA-small-subset.sorted.bam" />
            <param name="reference_source_selector" value="history" />
            <param name="ref_file" value="enolase_reference.fas" />
            <!-- <param name="w" value="tig00000001:200000-202000" /> -->
            <output name="polya_results" file="30xpolyA-small-subset-results.tsv" compare="sim_size">
                <assert_contents>
                    <has_n_lines n="14"/>
                    <has_n_columns n="10"/>
                    <has_line_matching expression="readname\scontig\sposition\sleader_start\sadapter_start\spolya_start\stranscript_start\sread_rate\spolya_length\sqc_tag"/>
                    <has_text text="453f3f3e-d22f-4d9c-81a6-8576e23390ed"/>
                    <has_text text="YHR174W"/>
                    <has_text text="READ_FAILED_LOAD"/>
                </assert_contents>
            </output>
        </test>
         <test>
            <param name="input_merged" ftype="fastq" value="30xpolyA-small-subset.fastq" />
            <param name="input_reads_raw" ftype="fast5.tar.gz" value="fast5_files_30xpolyA-small-subset.tar.gz" />
            <param name="b" value="30xpolyA-small-subset.sorted.bam" />
            <param name="reference_source_selector" value="history" />
            <param name="ref_file" value="enolase_reference.fas" />
            <param name="w" value="YHR174W:600-900" />
            <output name="polya_results" file="30xpolyA-small-subset-win-results.tsv" compare="sim_size">
                <assert_contents>
                    <has_n_lines n="12"/>
                    <has_n_columns n="10"/>
                    <has_line_matching expression="readname\scontig\sposition\sleader_start\sadapter_start\spolya_start\stranscript_start\sread_rate\spolya_length\sqc_tag"/>
                    <has_text text="453f3f3e-d22f-4d9c-81a6-8576e23390ed"/>
                    <has_text text="YHR174W"/>
                    <has_text text="READ_FAILED_LOAD"/>
                </assert_contents>
            </output>
        </test>
         <test>
            <param name="input_merged" ftype="fastq" value="30xpolyA-small-subset.fastq" />
            <param name="input_reads_raw" ftype="fast5.tar" value="fast5_files_30xpolyA-small-subset.tar" />
            <param name="b" value="30xpolyA-small-subset.sorted.bam" />
            <param name="reference_source_selector" value="history" />
            <param name="ref_file" value="enolase_reference.fas" />
            <param name="w" value="YHR174W:600-900" />
            <output name="polya_results" file="30xpolyA-small-subset-win-results-t3.tsv" compare="sim_size">
                <assert_contents>
                    <has_n_lines n="12"/>
                    <has_n_columns n="10"/>
                    <has_line_matching expression="readname\scontig\sposition\sleader_start\sadapter_start\spolya_start\stranscript_start\sread_rate\spolya_length\sqc_tag"/>
                    <has_text text="453f3f3e-d22f-4d9c-81a6-8576e23390ed"/>
                    <has_text text="YHR174W"/>
                    <has_text text="READ_FAILED_LOAD"/>
                </assert_contents>
            </output>
        </test>        
    </tests>    
    <help><![CDATA[
        Usage: nanopolish polya [OPTIONS] --reads reads.fa --bam alignments.bam --genome genome.fa 
        Estimate the length of the poly-A tail on direct RNA reads   

        Quickstart tutorial and manual available at:
        http://nanopolish.readthedocs.io/en/latest/quickstart_polya.html

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