annotate preprocessing.xml @ 11:c0c9d19bc7b2 draft

planemo upload for repository https://github.com/eteriSokhoyan/galaxytools/tree/branchForIterations/tools/GraphClust commit 746497a64b955f6b9afc1944d1c1d8d877e53267

<tool id="preproc" name="Preprocessing" version="0.3">
  <requirements>
    <requirement type="package" version="0.3.1">graphclust-wrappers</requirement>
    <requirement type="package" version="3.0">zip</requirement>
    <requirement type="package" version="1.70">biopython</requirement>

  </requirements>
    <stdio>
        <exit_code range="1:" />
    </stdio>
    <command>
		<![CDATA[
        preprocessing.pl
            '$fastaFile'
            $max_length
            $in_winShift
            $min_seq_length

        #if $SHAPEdata:
            &&
            python '$__tool_directory__/splitSHAPE.py' 
                '$SHAPEdata'
                
        #end if

        #if $AlignmentData:
            &&
            python '$__tool_directory__/splitStockholm.py' 
                '$AlignmentData'
                
        #end if
             
]]>
	</command>
    <inputs>
        <param type="data" name="fastaFile" format="fasta" />
        <param type="data" name="SHAPEdata" format="txt" optional="true" label="SHAPE data"/>
        <param type="data" name="AlignmentData" format="stockholm" optional="true" label="Alignments file"/>
        <param name="max_length" type="integer" value="10000" size="5" label="window size"/>
        <param name="in_winShift" type="integer" value="100" size="5" label="window shift in percent"/>
        <param name="min_seq_length" type="integer" value="5" size="5" label="minimum sequence length"/>
    </inputs>
    <outputs>
        <data name="data.fasta" format="fasta" from_work_dir="FASTA/data.fasta" label="data.fasta"/>
        <data name="data.map" format="txt" from_work_dir="FASTA/data.map" label="data.map"/>
        <data name="data.names" format="txt" from_work_dir="FASTA/data.names" label="data.names"/>
        <data name="data.fasta.scan" format="fasta" from_work_dir="FASTA/data.fasta.scan" label="data.fasta.scan"/>
        <data name="FASTA" format="zip" from_work_dir="FASTA.zip" label="FASTA.ZIP"/>
        <data name="shape_data_split" format="txt" from_work_dir="shape_data_split.react" label="SHAPE.data.split"/>
        <data name="alignment_data_split" format="stockholm" from_work_dir="alignment_data_split.stk" label="alignments.data.stk"/>
    </outputs>
    <tests>
        <test>
            <param name="fastaFile" value="input.fa"/>
            <param name="max_length" value="10000"/>
            <param name="in_winShift" value="100"/>
            <param name="min_seq_length" value="5"/>
            <output name="data.fasta" file="FASTA/data.fasta"/>
            <output name="data.map" file="FASTA/data.map" />
            <output name="data.names" file="FASTA/data.names"/>
            <output name="data.fasta.scan" file="FASTA/data.fasta.scan" />
        </test>

        <test>
            <param name="fastaFile" value="sample_3.fa"/>
            <param name="SHAPEdata" value="sample_3.react"/>
            <param name="max_length" value="100"/>
            <param name="in_winShift" value="50"/>
            <param name="min_seq_length" value="5"/>
            <output name="shape_data_split" file="sample_3_shape_data_split.react" />
        </test> 
        <test>
            <param name="fastaFile" value="sample_4_representatives.fa"/>
            <param name="AlignmentData" value="sample_4_all.stk"/>
            <param name="max_length" value="50"/>
            <param name="in_winShift" value="50"/>
            <param name="min_seq_length" value="5"/>
            <output name="alignment_data_split" file="sample_4_alignment_data_split.stk" />
        </test>        
    </tests>
    <help>
<![CDATA[

**What it does**

The tool takes as input a set of sequences in Fasta format and creates the final input for GraphCLust based on given parameters.

**Parameters**

+ **window size** : All input sequences are splitted into fragments of this length.
  The shift of the sliding window can be defined via option *window shift in percent*.
  This paramter reflects the expected length of signals to be found.
  Slightly larger windows are usually ok. Too small windows can disturb existing signals.


+ **window shift in percent** : Relative window size in % for window shift during input preprocessing.
  Please note that a small shift results in much more fragments for clustering. The benefit is that RNA
  motifs/structures are not destroyed by arbitrary split points. Smaller
  shifts usually increase the cluster quality. Too small shifts (<20) are not
  recommended as a dense center is "polluted" by overlapping fragments and
  no other occurences in the dataset can be found.


+ **minimum sequence length** : Minimal length of input sequences.
  Every input sequence below that length is ignored completely during clustering.

    ]]></help>
    <citations>
        <citation type="doi">10.1093/bioinformatics/bts224</citation>
    </citations>
</tool>