annotate abims_hclustering.xml @ 0:2f7381ee5235 draft

Uploaded

<tool id="abims_hclustering" name="Hierarchical Clustering" version="1.1">

    <description>using ctc R package for java-treeview</description>

    <command interpreter="Rscript">
        abims_hclustering.r file "$input" method $method link $link keep.hclust FALSE normalization $normalization sep "$sep" dec "$dec"
    </command>

    <inputs>
        <param name="input" type="data" label="Data Matrix file" format="tabular" help="Matrix of numeric data with headers." />
        <param name="method" type="select" label="Distance measure method" help="the distance measure to be used">
            <option value="pearson" selected="true">pearson</option>
            <option value="euclidean" >euclidean</option>
            <option value="maximum" >maximum</option>
            <option value="manhattan" >manhattan</option>
            <option value="canberra" >canberra</option>
            <option value="binary" >binary</option>
            <option value="correlation" >correlation</option>
            <option value="spearman" >spearman</option>
        </param>
        <param name="link" type="select" label="Agglomeration/Link method" help="the agglomeration method to be used">
            <option value="ward" selected="true">ward</option>
            <option value="single" >single</option>
            <option value="complete" >complete</option>
            <option value="average" >average</option>
            <option value="mcquitty" >mcquitty</option>
            <option value="median" >median</option>
            <option value="centroid" >centroid</option>
        </param>
        <param name="normalization" type="select" label="Normalization by center and scale" help="Centering is done by subtracting the column means and scaling is done by dividing the (centered) columns of by their standard deviations">
            <option value="T" selected="true">TRUE</option>
            <option value="F" >FALSE</option>
        </param>
        
        <param name="sep" type="select" format="text" optional="true">
		<label>Separator of columns</label>
		<option value="tabulation">tabulation</option>
		<option value="semicolon">;</option>
		<option value="comma">,</option>
	</param>
        <param name="dec" type="text" label="Decimal separator" value="." help="" />
        
        <!--<param name="nr_col_names" type="integer" label="names" value="2" help="number of the column with names of metabolits" />
        <param name="from" type="integer" label="from" value="15" help="number of the column starting peak values data (to exlude all metadata)" />
        <param name="to" type="integer" label="to" value="30" help="number of the column finishing peak values data (to exlude all metadata)" />
        <param name="gr_number" type="integer" label="gr_number" value="2" help="number of groups (conditions)" />
        <param name="nb_col_gr" type="text" label="nb_col_gr" value="8,8" help="number of column of each group; separate with coma as indicated; first position coresponding to the first group etc." />
        <param name="threshold" type="float" label="threshold" value="0.01" help="max adjusted p.value accepted" />-->
        
    </inputs>

    <outputs>
        <data name="hclust_zip" format="zip" from_work_dir="hclust.zip" label="${input.name[:-4]}.hclust.zip" />
    </outputs>

    <stdio>
        <exit_code range="1:" level="fatal" />
    </stdio>

    <help>


		
.. class:: infomark

**Authors** Gildas Le Corguille  ABiMS - UPMC/CNRS - Station Biologique de Roscoff - gildas.lecorguille|at|sb-roscoff.fr 

---------------------------------------------------

=======================
Hierarchical Clustering
=======================

-----------
Description
-----------

This function compute hierachical clustering with function
hcluster and export cluster to Java TreeView files format: jtreeview.sourceforge.net.

This function performs a **hierarchical cluster analysis** using a set
of dissimilarities for the n objects being clustered.  Initially,
each object is assigned to its own cluster and then the algorithm
proceeds iteratively, at each stage joining the two most similar
clusters, continuing until there is just a single cluster.  At
each stage distances between clusters are recomputed by the
Lance-Williams dissimilarity update formula according to the
particular clustering method being used.

A number of different **clustering methods** are provided.  **Ward's**
minimum variance method aims at finding compact, spherical
clusters.  The **complete linkage** method finds similar clusters.
The **single linkage** method (which is closely related to the
minimal spanning tree) adopts a ‘friends of friends’ clustering
strategy.  The other methods can be regarded as aiming for
clusters with characteristics somewhere between the single and
complete link methods.  Note however, that methods **median** and
**centroid** are not leading to a monotone distance measure,
or equivalently the resulting dendrograms can have so called
inversions (which are hard to interpret).




-----------
Input files
-----------

+---------------------------+------------+
| Parameter : num + label   |   Format   |
+===========================+============+
| 1 : Data Matrix file      | Tabular    |
+---------------------------+------------+


----------
Parameters
----------


**Agglomeration or Link method:*

A number of different clustering methods are provided. Ward's minimum variance method aims at finding compact, spherical clusters.
The complete linkage method finds similar clusters. The single linkage method (which is closely related to the minimal spanning tree) adopts a ‘friends of friends’ clustering strategy.
The other methods can be regarded as aiming for clusters with characteristics somewhere between the single and complete link methods.
Note however, that methods median and centroid are not leading to a monotone distance measure, or equivalently the resulting dendrograms can have so called inversions (which are hard to interpret).


------------
Output files
------------

***.tab.hclust.zip**

	| A zip file containing three files (hclust.atr, hclust.cdt and hclust.gtr) that are Treeview format. If you want to have more informations or download Treeview, you can visit the webiste:
	| http://jtreeview.sourceforge.net
	

	
------

.. class:: infomark 

You can continue your analysis using Treeview (outside of Galaxy) with the three files (atr,cdt and gtr) within the **xset.tab.hclust.zip** output.
	

	
	
---------------------------------------------------

---------------
Working example
---------------


Input files
-----------

**>A part of an example of Data Matrix file input**
	

+--------+------------------+----------------+
| Name   | Bur-eH_FSP_102   | Bur-eH_FSP_22  |	                                       
+========+==================+================+
|M202T601| 91206595.7559783 |106808979.08546 |
+--------+------------------+----------------+
|M234T851| 27249137.275504  |28824971.3177926|  
+--------+------------------+----------------+  


Parameters
----------

	| Distance measure method  -> **pearson**
	| Agglomeration/Link method -> **ward**
	| Normalization by center and scale -> **TRUE**
	| Separator of columns -> **tabulation**
	| Decimal separator: -> **.**
	


Output files
------------

**Example of an dendrogram/heatmap generated by the Treeview tool**:

.. image:: hclust.png


    </help>

</tool>