view filter_assembly.xml @ 0:7a813e633d1c draft default tip

planemo upload for repository https://github.com/abims-sbr/adaptsearch commit 3c7982d775b6f3b472f6514d791edcb43cd258a1-dirty

<tool name="Filter assemblies" id="filter_assemblies" version="2.0.3">

    <description>
        Filter the outputs of Velvet or Trinity assemblies
    </description>

    <macros>
        <import>macros.xml</import>
    </macros>

    <requirements>
        <expand macro="python_required" />
        <requirement type="package" version="0.0.14">fastx_toolkit</requirement>
        <requirement type="package" version="10.2011">cap3</requirement>
    </requirements>

    <command>
    <![CDATA[
        #set $infiles = ""
        #for $input in $inputs
            ln -s '$input' '$input.element_identifier';
            #set $infiles = $infiles + $input.element_identifier + ","
        #end for
        #set $infiles = $infiles[:-1]

        ln -s '$__tool_directory__/scripts/S02a_remove_redondancy_from_velvet_oases.py' . &&
        ln -s '$__tool_directory__/scripts/S02b_format_fasta_name_trinity.py' . &&
        ln -s '$__tool_directory__/scripts/S03_choose_one_variants_per_locus_trinity.py' . &&
        ln -s '$__tool_directory__/scripts/S04_find_orf.py' . &&
        ln -s '$__tool_directory__/scripts/S05_filter.py' . &&

        python '$__tool_directory__/scripts/S01_script_to_choose.py'

        '$infiles'
        $length_seq_max
        $percent_identity
        $overlap_length
        > ${log}
    ]]>
    </command>

    <inputs>
        <param name="inputs" type="data" format="fasta" multiple="true" label="Input files" />
        <param name="percent_identity" type="integer" value="100" label="Overlap percent identity cutoff" help="Cap3 parameter (-p N); minimum percent identity of an overlap. The specified value should be more than 65%." />
        <param name="overlap_length" type="integer" value="60" label="Overlap length cutoff" help="Cap3 parameter (-o N); minimum length of an overlap (in base pairs). The specified value should be more than 15 base pairs." />
        <param name="length_seq_max" type="integer" value="100" label="Minimum sequence length" help="Keep sequences which length is higher than the minimum sequence length  " />
    </inputs>

    <outputs>
        <collection name="output_fasta" type="list" label="Filter Assemblies outputs">
            <discover_datasets pattern="__name_and_ext__" directory="outputs" />
        </collection>
        <data format="txt" name="log" label="Filter Assemblies Summary"/>
    </outputs>

	<tests>
        <test>
            <param name="inputs" ftype="fasta" value="trinity/Pfiji_trinity.fasta,trinity/Apomp_trinity.fasta,trinity/Amphi_trinity.fasta,trinity/Acaud_trinity.fasta,velvet/Pg_transcriptome_90109.fasta,velvet/Ap_transcriptome_35099.fasta,velvet/Ac_transcriptome_25591.fasta" />
            <param name="percent_identity" value="100" />
            <param name="overlap_length" value="60" />
            <param name="length_seq_max" value="100" />
            <output name="log" value="trinity_and_velvet_up.output" />
            <output_collection name="output_fasta" type="list">
                <element name="AcAc_transcriptome_25591" value="velvet_out/AcAc_transcriptome_25591.fasta" />
                <element name="ApAp_transcriptome_35099" value="velvet_out/ApAp_transcriptome_35099.fasta" />
                <element name="PgPg_transcriptome_90109" value="velvet_out/PgPg_transcriptome_90109.fasta" />
                <element name="AcAcaud_trinity" value="trinity_out/AcAcaud_trinity.fasta" />
                <element name="AmAmphi_trinity" value="trinity_out/AmAmphi_trinity.fasta" />
                <element name="ApApomp_trinity" value="trinity_out/ApApomp_trinity.fasta" />
                <element name="PfPfiji_trinity" value="trinity_out/PfPfiji_trinity.fasta" />
            </output_collection>
        </test>        
        <test>
            <param name="inputs" ftype="fasta" value="trinity/Pfiji_trinity.fasta,trinity/Apomp_trinity.fasta,trinity/Amphi_trinity.fasta,trinity/Acaud_trinity.fasta" />
            <param name="percent_identity" value="100" />
            <param name="overlap_length" value="60" />
            <param name="length_seq_max" value="100" />
            <output name="log" value="trinity_up.output" />
            <output_collection name="output_fasta" type="list">
                <element name="AcAcaud_trinity" value="trinity_out/AcAcaud_trinity.fasta" />
                <element name="AmAmphi_trinity" value="trinity_out/AmAmphi_trinity.fasta" />
                <element name="ApApomp_trinity" value="trinity_out/ApApomp_trinity.fasta" />
                <element name="PfPfiji_trinity" value="trinity_out/PfPfiji_trinity.fasta" />
            </output_collection>
        </test>
        <test>
            <param name="inputs" ftype="fasta" value="velvet/Pg_transcriptome_90109.fasta,velvet/Ap_transcriptome_35099.fasta,velvet/Ac_transcriptome_25591.fasta" />
            <param name="percent_identity" value="100" />
            <param name="overlap_length" value="60" />
            <param name="length_seq_max" value="100" />
            <output name="log" value="velvet_up.output" />
            <output_collection name="output_fasta" type="list">
                <element name="AcAc_transcriptome_25591" value="velvet_out/AcAc_transcriptome_25591.fasta" />
                <element name="ApAp_transcriptome_35099" value="velvet_out/ApAp_transcriptome_35099.fasta" />
                <element name="PgPg_transcriptome_90109" value="velvet_out/PgPg_transcriptome_90109.fasta" />
            </output_collection>
        </test>
        
    </tests>

	<help>

@HELP_AUTHORS@

<![CDATA[

**Description**

This tool reformats Velvet Oases or Trinity assemblies for the AdaptSearch galaxy suite and selects only one variant per gene according to its length and quality check.

---------

**Input format**

(1) Sequences are in the sequential format:

| >seqname1
| AAAGAGAGACCACATGTCAGTAGC -on one or several lines -
| >seqname2
| AAGGCCTGACCACATGAGTTAAGC -on one or several lines -
| etc ...
|

2) The file name should begin with a two letter abbreviation of the species name (for isntance, 'Ap' if the species is Alvinella pompejana).

**For Velvet Oases assemblies input**
            
    The headers must be as follow : *>Locus_i_Transcript_i/j_Confidence_x.xxx_Length_N* where i is the locus number, j the transcript variant among all versions of the transcript, x.xxx the confidence value and N the length.

**For Trinity assemblies inputs**   
            
    The headers must be as follow : *>cj_gj_ij Len=j path=[j:0-j]* where all the j are integers (locus number, transcript variant, length, position...)

**The tool handles the case if input files come from both assemblers (there is no need for input files to be exclusively from one or another assembler).**

---------

**Parameters**

    - 'Input files' : a collection of fasta files (one file per species).
    - 'Overlap percent identity cutoff' : cap3 -p parameter : minimum percent identity of an overlap.
        must be > 65 ; default : 100.
    - 'Overlap length cutoff' (integer) : cap3 -o parameter : minimum length of an overlap (in base pairs).
        must be > 15 ; default : 60.
    - 'Minimum sequence length' (integer) : only keep sequences which are longer than the specified value.
        default : 100.

---------

**Steps**:
    
The tool:
    1) Modifies the sequence name to add the species abbreviation using the 2 first letters of the name of the transcriptome file : note that each species abbreviation must be unique
    2) Selects one allelic sequence from each transcript (c or locus) using the length of the sequence and its level of confidence
    3) Selects the best ORF from the sequence between two stop codons
    4) Performs a CAP3 from the full set of ORFs to minimize redundancy
    5) Retrieves the initial transcript sequences from the remaining set of proceeded ORF sequences

**Outputs**

    - 'Filter Assemblies Summary' : the log file.
    - 'Filter Assemblies outputs' : the main results.

---------

**The AdaptSearch Pipeline**

.. image:: adaptsearch_picture_helps.png

---------

Changelog
---------

**Version 2.1 - 15/01/2018**

    - Input files can be a mix from files coming either from Trinity or Velvet Oases assemblers

**Version 2.0 - 14/04/2017**

    - NEW: Replace the zip between tools by Dataset Collection

**Version 1.0 - 13/04/2017**

    - TEST: Add funtional test with planemo
    - IMPROVEMENT: Use conda dependencies for cap3, fastaformatter and python

    ]]>
	</help>

</tool>