comparison cluster.xml @ 4:05696474ee89 draft

planemo upload for repository https://github.com/galaxyproject/tools-devteam/tree/master/tool_collections/gops/cluster commit cae3e05d02e60f595bb8b6d77a84f030e9bd1689

3:765ceb06c3e2

<tool id="gops_cluster_1" name="Cluster" version="1.0.0">
  <description>the intervals of a dataset</description>
  <requirements>
    <requirement type="package" version="0.7.1">bx-python</requirement>
    <requirement type="package" version="1.0.0">galaxy-ops</requirement>
  </requirements>
  <trackster_conf/>
  <command interpreter="python">gops_cluster.py $input1 $output -1 ${input1.metadata.chromCol},${input1.metadata.startCol},${input1.metadata.endCol},${input1.metadata.strandCol} -d $distance -m $minregions -o $returntype</command>
  <inputs>
    <param format="interval" name="input1" type="data">
      <label>Cluster intervals of</label>
    </param>
    <param name="distance" type="integer" value="1" help="(bp)">
      <label>max distance between intervals</label>
    </param>
    <param name="minregions" type="integer" value="2">
      <label>min number of intervals per cluster</label>
    </param>
	<param name="returntype" type="select" label="Return type">
		<option value="1">Merge clusters into single intervals</option>
		<option value="2">Find cluster intervals; preserve comments and order</option>
		<option value="3">Find cluster intervals; output grouped by clusters</option>
		<option value="4">Find the smallest interval in each cluster</option>
		<option value="5">Find the largest interval in each cluster</option>
	</param>
   </inputs>
  <outputs>
    <data format="input" name="output" metadata_source="input1" />
  </outputs>
  <code file="operation_filter.py">
    <hook exec_after_process="exec_after_cluster" />
  </code>
  <tests>
    <test>
      <param name="input1" value="5.bed" />
      <param name="distance" value="1" />
      <param name="minregions" value="2" />
      <param name="returntype" value="1" />
      <output name="output" file="gops-cluster-1.bed" />     
    </test>
    <test>
      <param name="input1" value="gops_cluster_bigint.bed" />
      <param name="distance" value="1" />
      <param name="minregions" value="2" />
      <param name="returntype" value="1" />
      <output name="output" file="gops-cluster-1.bed" />     
    </test>
    <test>
      <param name="input1" value="5.bed" />
      <param name="distance" value="1" />
      <param name="minregions" value="2" />
      <param name="returntype" value="2" />
      <output name="output" file="gops-cluster-2.bed" />     
    </test>    
    <test>
      <param name="input1" value="5.bed" />
      <param name="distance" value="1" />
      <param name="minregions" value="2" />
      <param name="returntype" value="3" />
      <output name="output" file="gops-cluster-3.bed" />     
    </test>
  </tests>
  <help>

.. class:: infomark

**TIP:** If your dataset does not appear in the pulldown menu, it means that it is not in interval format. Use "edit attributes" to set chromosome, start, end, and strand columns.

-----

**Screencasts!**

See Galaxy Interval Operation Screencasts_ (right click to open this link in another window).

.. _Screencasts: http://wiki.g2.bx.psu.edu/Learn/Interval%20Operations

-----

**Syntax**

- **Maximum distance** is greatest distance in base pairs allowed between intervals that will be considered &quot;clustered&quot;.  **Negative** values for distance are allowed, and are useful for clustering intervals that overlap.
- **Minimum intervals per cluster** allow a threshold to be set on the minimum number of intervals to be considered a cluster.  Any area with less than this minimum will not be included in the output.

.. image:: gops_clusterFind.gif

Merge Clusters:

.. image:: gops_clusterMerge.gif

</help>
</tool>

4:05696474ee89

<tool id="gops_cluster_1" name="Cluster" version="1.0.0">
    <description>the intervals of a dataset</description>
    <macros>
        <import>macros.xml</import>
    </macros>
    <expand macro="requirements" />
    <code file="operation_filter.py">
        <hook exec_after_process="exec_after_cluster" />
    </code>
    <command><![CDATA[
python '$__tool_directory__/gops_cluster.py'
'$input1'
'$output'
-1 ${input1.metadata.chromCol},${input1.metadata.startCol},${input1.metadata.endCol},${input1.metadata.strandCol}
-d $distance
-m $minregions
-o $returntype
    ]]></command>
    <inputs>
        <param name="input1" type="data" format="interval" label="Cluster intervals of" />
        <param name="distance" type="integer" value="1" label="Max distance between intervals" help="(bp)" />
        <param name="minregions" type="integer" value="2" label="Min number of intervals per cluster" />
        <param name="returntype" type="select" label="Return type">
            <option value="1">Merge clusters into single intervals</option>
            <option value="2">Find cluster intervals; preserve comments and order</option>
            <option value="3">Find cluster intervals; output grouped by clusters</option>
            <option value="4">Find the smallest interval in each cluster</option>
            <option value="5">Find the largest interval in each cluster</option>
        </param>
    </inputs>
    <outputs>
        <data name="output" format_source="input1" metadata_source="input1" />
    </outputs>
    <tests>
        <test>
            <param name="input1" value="5.bed" />
            <param name="distance" value="1" />
            <param name="minregions" value="2" />
            <param name="returntype" value="1" />
            <output name="output" file="gops-cluster-1.bed" />
        </test>
        <test>
            <param name="input1" value="gops_cluster_bigint.bed" />
            <param name="distance" value="1" />
            <param name="minregions" value="2" />
            <param name="returntype" value="1" />
            <output name="output" file="gops-cluster-1.bed" />
        </test>
        <test>
            <param name="input1" value="5.bed" />
            <param name="distance" value="1" />
            <param name="minregions" value="2" />
            <param name="returntype" value="2" />
            <output name="output" file="gops-cluster-2.bed" />
        </test>
        <test>
            <param name="input1" value="5.bed" />
            <param name="distance" value="1" />
            <param name="minregions" value="2" />
            <param name="returntype" value="3" />
            <output name="output" file="gops-cluster-3.bed" />
        </test>
    </tests>
    <help><![CDATA[
.. class:: infomark

**TIP:** If your dataset does not appear in the pulldown menu, it means that it is not in interval format. Use "edit attributes" to set chromosome, start, end, and strand columns.

@SCREENCASTS@

**Syntax**

- **Maximum distance** is greatest distance in base pairs allowed between intervals that will be considered &quot;clustered&quot;.  **Negative** values for distance are allowed, and are useful for clustering intervals that overlap.
- **Minimum intervals per cluster** allow a threshold to be set on the minimum number of intervals to be considered a cluster.  Any area with less than this minimum will not be included in the output.

.. image:: gops_clusterFind.gif

Merge Clusters:

.. image:: gops_clusterMerge.gif
    ]]></help>
</tool>