annotate searchgui.xml @ 59:943a34df6047 draft

"planemo upload for repository https://github.com/galaxyproteomics/tools-galaxyp/tree/master/tools/peptideshaker commit 1cb321cbce2a6528764abc9c9431687f56c57588"

<tool id="search_gui"
      name="Search GUI"
      version="@SEARCHGUI_VERSION@+galaxy@SEARCHGUI_VERSION_SUFFIX@"
      profile="20.01"
>
    <description>
        Perform protein identification using various search engines and prepare results for input to Peptide Shaker
    </description>
    <macros>
        <import>macros_basic.xml</import>
    </macros>
    <requirements>
        <requirement type="package" version="@SEARCHGUI_VERSION@">searchgui</requirement>
        <requirement type="package" version="3.0">zip</requirement>
    </requirements>
    <expand macro="stdio" />
    <command use_shared_home="false">

<![CDATA[
        #from datetime import datetime
        #import json
        #import os
        #set $exp_str = "Galaxy_Experiment_%s" % datetime.now().strftime("%Y%m%d%H%M%s")
        #set $samp_str = "Sample_%s" % datetime.now().strftime("%Y%m%d%H%M%s")
        #set $bin_dir = "bin"

        mkdir output;
        mkdir output_reports;
        mkdir temp_folder;
        mkdir log_folder;
        trap ">&2 cat log_folder/*.log" EXIT;

        cwd=`pwd`;
        export HOME=\$cwd;

        ## echo the search engines to run (single quotes important because X!Tandem)
        echo '$search_engines_options.engines';
        echo 'DB: ${input_fasta_file.element_identifier} sequences: ${input_fasta_file.metadata.sequences}';

        ##Create a searchgui.properties file for the version, which will be added to the searchgui_results if not already present
        echo 'searchgui.version=@SEARCHGUI_VERSION@' >> searchgui.properties;

        #for $peak_list_file in $peak_lists_files:
            #set $input_name = ""
            #if $peak_list_file.is_of_type("mgf"):
                #set $input_name = $peak_list_file.element_identifier.split('/')[-1].replace(".mgf", "") + ".mgf"
            #else if $peak_list_file.is_of_type("mzml"):
                #set $input_name = $peak_list_file.element_identifier.split('/')[-1].replace(".mzml", "") + ".mzml"
            #end if
            ln -s -f '${peak_list_file}' '${input_name}';
            #set $encoded_id = $__app__.security.encode_id($peak_list_file.id)
            echo 'Spectrums:${peak_list_file.element_identifier}(API:${encoded_id})';
        #end for

        ## copy the input .par file to the working folder
        cp '${input_parameters_file}' './SEARCHGUI_IdentificationParameters.par';

        ## copy the input .fasta file to the working folder
        cp '${input_fasta_file}' './input_fasta_file.fasta';


        ################
        ## Search CLI ##
        ################
        echo 'running search gui' &&
        searchgui -Djava.awt.headless=true eu.isas.searchgui.cmd.SearchCLI
            --exec_dir="\$cwd/${bin_dir}"
            -use_log_folder 0
            -spectrum_files \$cwd
            -fasta_file "\$cwd/input_fasta_file.fasta"
            -output_folder \$cwd/output
            -id_params ./SEARCHGUI_IdentificationParameters.par
            -temp_folder \$cwd/temp_folder
            -log \$cwd/log_folder
            -threads "\${GALAXY_SLOTS:-12}"

            #if $searchgui_advanced.searchgui_advanced_selector == 'advanced'
                -correct_titles "${searchgui_advanced.correct_titles}"
                $searchgui_advanced.missing_titles
                -mgf_splitting "${searchgui_advanced.mgf_splitting}"
                -mgf_spectrum_count "${searchgui_advanced.mgf_spectrum_count}"
                -output_gzip "${searchgui_advanced.output_gzip}"
            #end if

            #set $engines_list = str($search_engines_options.engines).split(',')
            #if 'X!Tandem' in $engines_list:
                -xtandem 1
            #else
                -xtandem 0
            #end if

            #if 'MyriMatch' in $engines_list:
                -myrimatch 1
            #else
                -myrimatch 0
            #end if

            #if 'MSGF' in $engines_list:
                -msgf 1
            #else
                -msgf 0
            #end if

            #if 'OMSSA' in $engines_list:
                -omssa 1
            #else
                -omssa 0
            #end if

            #if 'Comet' in $engines_list:
                -comet 1
            #else
                -comet 0
            #end if

            #if 'Tide' in $engines_list:
                -tide 1
            #else
                -tide 0
            #end if

            #if 'MS_Amanda' in $engines_list:
                -ms_amanda 1
            #else
                -ms_amanda 0
            #end if

            #if 'Andromeda' in $engines_list:
                -andromeda 1
            #else
                -andromeda 0
            #end if

            #if 'MetaMorpheus' in $engines_list:
                -meta_morpheus 1
            #else
                -meta_morpheus 0
            #end if

            #if 'Novor' in $engines_list:
                -novor 1
            #else
                -novor 0
            #end if

            #if 'DirecTag' in $engines_list:
                -directag 1
            #else
                -directag 0
            #end if

            ## single zip file
            -output_option 0

            ## mgf and database in output
            -output_data 1

        &&

        echo '>log_folder:' &&

        ls -l \$cwd/log_folder &&

        mv output/searchgui_out.zip searchgui_out.zip

        &&

        zip -u searchgui_out.zip searchgui.properties
]]>
    </command>
    <inputs>
        <param format="json" name="input_parameters_file" type="data" label="Identification Parameters file"/>

        <param format="fasta" name="input_fasta_file" type="data" label="Fasta file" help="For postprocessing with PeptideShaker the sequences must have UniProt style headers and contain decoy sequences (see help)."/>

        <param name="peak_lists_files" format="mgf,mzml" type="data" multiple="true" label="Input Peak Lists"
            help="Select appropriate MGF/MZML dataset(s) from history.
               Remember to choose only .mzML files when using MetaMorpheus." />

        <!-- Search Engine Selection -->
        <section name="search_engines_options" expanded="true" title="Search Engine Options">
            <param name="engines" type="select" display="checkboxes" multiple="True" label="DB-Search Engines">
                <help>Comet and Tide shouldn't both be selected since they use a similar algoritm. OMSSA may not work into isolated environments like containers. 
                    MetaMorpheus only produce results when using mzML format.</help>
                <option value="X!Tandem" selected="True">X!Tandem</option>
                <option value="MSGF" selected="True">MS-GF+</option>
                <option value="OMSSA">OMSSA</option>
                <option value="Comet">Comet</option>
                <option value="Tide">Tide</option>
                <option value="MyriMatch">MyriMatch</option>
                <option value="MS_Amanda">MS_Amanda</option>
                <option value="MetaMorpheus">MetaMorpheus</option>
                <!-- Windows only
                <option value="Andromeda">Andromeda</option>
                -->
                <!-- New with version 3.0
                -->
                <!--working in tests
                -->
                <option value="DirecTag">DirecTag</option>
                <option value="Novor">Novor (Select for non-commercial use only)</option>
                <validator type="no_options" message="Please select at least one output file" />
            </param>
        </section>

        <conditional name="searchgui_advanced">
            <param name="searchgui_advanced_selector" type="select" label="SearchGUI Options">
                <option value="basic" selected="True">Default</option>
                <option value="advanced">Advanced</option>
            </param>
            <when value="basic" />
            <when value="advanced">
                <param name="correct_titles" type="select" label="How should PeptideShaker deal with duplicate spectra?"
                    help="Unless you suspect some input files to be genuine duplicates then rename spectra is the safest option">
                    <option value="0">no correction</option>
                    <option value="1" selected="True">rename spectra</option>
                    <option value="2">delete spectra</option>
                </param>
                <param name="missing_titles" type="boolean" checked="false" truevalue="-missing_titles 1" falsevalue="-missing_titles 0"
                    label="Add missing spectrum titles" help="(-missing_titles)"/>
                <param name="mgf_splitting" type="integer" value="1000" label="The maximum mgf file size in MB before splitting the mgf"
                    help="Choose a smaller value if you are running on a machine with limited memory"/>
                <param name="mgf_spectrum_count" type="integer" value="25000" label="The maximum number of spectra per mgf file when splitting"
                    help="Choose a smaller value if you are running on a machine with limited memory"/>
                <param name="output_gzip" type="boolean" checked="true" truevalue="1" falsevalue="0"
                    label="Gzip result files"/>
            </when>
        </conditional>

    </inputs>
    <outputs>
        <data name="searchgui_results" format="searchgui_archive" from_work_dir="searchgui_out.zip" label="${tool.name} on ${on_string}" />
    </outputs>
    <tests>

        <!-- Test that specifying non-default search engines with default parameters works -->
        <test>
            <param name="peak_lists_files" value="searchgui_tinyspectra1.mgf"/>
            <param name="input_parameters_file" value="Identification_Parameters_default.par"/>
            <param name="input_fasta_file" value="fastacli_searchgui_tinydb1_concatenated_target_decoy.fasta" ftype="fasta" />
            <param name="engines" value="X!Tandem,MSGF,MyriMatch,Comet"/>
            <output name="searchgui_results" file="searchgui_tiny_result_default_4engines.zip" ftype="searchgui_archive" compare="sim_size" delta="30000" />
        </test>
        
        
        <!-- Test that search works with MetaMorpheus with default parameters works-->
        <!-- Test data has been taken from metamorpheus galaxy tool -->
        <test>
            <param name="peak_lists_files" value="searchgui_smallCalibratible_Yeast.mzML"/>
            <param name="input_parameters_file" value="Identification_Parameters_default.par"/>
            <param name="input_fasta_file" value="searchgui_smallYeast.fasta" ftype="fasta" />
            <param name="engines" value="MetaMorpheus"/>
            <output name="searchgui_results" ftype="searchgui_archive">
                <assert_contents>
                    <has_size value="798597" delta="10000"/>
                </assert_contents>
            </output>
        </test>

        <!-- Test that search works with MSAmanda with default parameters works-->
        <test>
            <param name="peak_lists_files" value="searchgui_smallspectra.mgf"/>
            <param name="input_parameters_file" value="Identification_Parameters_default.par"/>
            <param name="input_fasta_file" value="fastacli_searchgui_tinydb1_concatenated_target_decoy.fasta" ftype="fasta" />
            <param name="engines" value="MS_Amanda"/>
            <output name="searchgui_results" ftype="searchgui_archive">
                <assert_contents>
                    <has_size value="635138" delta="5000"/>
                </assert_contents>
            </output>
        </test>

		    <!-- Test that specifying non-default search engines with non-default parameters works -->
        <test>
           <param name="peak_lists_files" value="searchgui_tinyspectra1.mgf"/>
           <param name="input_parameters_file" value="Identification_Parameters_specific.par"/>
           <param name="input_fasta_file" value="fastacli_searchgui_tinydb1_concatenated_target_decoy.fasta" ftype="fasta" />
           <param name="engines" value="X!Tandem,MSGF,MyriMatch,Comet"/>
           <output name="searchgui_results" ftype="searchgui_archive">
               <assert_contents>
                   <has_size value="159330" delta="20000"/>
               </assert_contents>
           </output>
        </test>

        <!--
            NOTE: Identification_Parameters_specific.par is equivalent to the default .par plus these parameters:
            <param name="precursor_ion_tol" value="100"/>
            <param name="min_charge" value="1"/>
            <param name="max_charge" value="3"/>
            <param name="xtandem|xtandem_advanced" value="yes"/>
            <param name="xtandem|xtandem_advanced|xtandem_refine_selector" value="yes"/>
        -->

        <!-- Test that specifying MsAmanda as search engine with non-default parameters works -->
        <test>
            <param name="peak_lists_files" value="searchgui_smallspectra.mgf"/>
            <param name="input_parameters_file" value="Identification_Parameters_specific.par"/>
            <param name="input_fasta_file" value="fastacli_searchgui_tinydb1_concatenated_target_decoy.fasta" ftype="fasta" />
            <param name="engines" value="MS_Amanda"/>
            <output name="searchgui_results" ftype="searchgui_archive">
                <assert_contents>
                    <has_size value="635158" delta="5000"/>
                </assert_contents>
            </output>
        </test>

        <!-- Test that specifying non-default search engines with default parameters works using modifications -->
        <test>
            <param name="peak_lists_files" value="searchgui_tinyspectra1.mgf"/>
            <param name="input_parameters_file" value="Identification_Parameters_default_modifications.par"/>
            <param name="input_fasta_file" value="fastacli_searchgui_tinydb1_concatenated_target_decoy.fasta" ftype="fasta" />
            <param name="engines" value="X!Tandem,MSGF,MyriMatch,Comet"/>
            <output name="searchgui_results" file="searchgui_tiny_result_default_4engines_modifications.zip" ftype="searchgui_archive" compare="sim_size" delta="30000" />
        </test>

        <!--
            NOTE: Identification_Parameters_default_modifications.par adds:
            - "Carbamidomethylation of C" as fixed modification
            - "Oxidation of M" as variable modification
          -->

        <!-- Test that search works with MSAmanda with default default parameters - with modifications -->
        <test>
            <param name="peak_lists_files" value="searchgui_tinyspectra1.mgf"/>
            <param name="input_parameters_file" value="Identification_Parameters_default_modifications.par"/>
            <param name="input_fasta_file" value="fastacli_searchgui_tinydb1_concatenated_target_decoy.fasta" ftype="fasta" />
            <param name="engines" value="MS_Amanda"/>
            <output name="searchgui_results" ftype="searchgui_archive">
                <assert_contents>
                    <has_size value="118136" delta="30000"/>
                </assert_contents>
            </output>
        </test>

    </tests>
    <help>
**What it does**

Runs multiple search engines on any number of MGF peak lists using SearchGUI.

Default:     X! Tandem and MS-GF+ are executed.

Optional:     MyriMatch, MS-Amanda, MetaMorpheus, OMSSA (it may not work into isolated environments like containers), Comet, Tide, DirecTag and Novor can be executed.

**Input FASTA**

In order to allow for postprocessing with PeptideShaker the sequences must contain decoy sequences (see _here: https://github.com/compomics/searchgui/wiki/DatabaseHelp#decoy-sequences)
and the FASTA header must either contain no "|" characters (then the whole header will be used as ID) or have the following format:

    >generic[your tag]|[protein accession]|[protein description]

    or 

    >generic[your tag]|[protein accession]

See _here: https://github.com/compomics/searchgui/wiki/DatabaseHelp#non-standard-fasta.



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