view tools/stats/grouping.xml @ 2:c2a356708570

Uploaded
author xuebing
date Fri, 09 Mar 2012 19:45:42 -0500
parents 9071e359b9a3
children
line wrap: on
line source

<tool id="Grouping1" name="Group" version="2.0.0">
  <description>data by a column and perform aggregate operation on other columns.</description>
  <command interpreter="python">
    grouping.py 
      $out_file1
      $input1
      $groupcol
      $ignorecase
      #for $op in $operations
       '${op.optype}
        ${op.opcol}
        ${op.opround}'
      #end for
  </command>
  <inputs>
    <param format="tabular" name="input1" type="data" label="Select data" help="Dataset missing? See TIP below."/>
    <param name="groupcol" label="Group by column" type="data_column" data_ref="input1" />
    <param name="ignorecase" type="boolean" truevalue="1" falsevalue="0">
      <label>Ignore case while grouping?</label>
    </param>
    <repeat name="operations" title="Operation">
      <param name="optype" type="select" label="Type">
        <option value="mean">Mean</option>
        <option value="median">Median</option>
        <option value="mode">Mode</option>
        <option value="max">Maximum</option>
        <option value="min">Minimum</option>
        <option value="sum">Sum</option>
        <option value="length">Count</option>
        <option value="unique">Count Distinct</option>
        <option value="cat">Concatenate</option>
        <option value="cat_uniq">Concatenate Distinct</option>
        <option value="random">Randomly pick</option>
        <option value="std">Standard deviation</option>
      </param>
      <param name="opcol" label="On column" type="data_column" data_ref="input1" />
      <param name="opround" type="select" label="Round result to nearest integer?">
         <option value="no">NO</option>
         <option value="yes">YES</option>
       </param>
    </repeat>
  </inputs>
  <outputs>
    <data format="tabular" name="out_file1" />
  </outputs>
  <requirements>
    <requirement type="python-module">numpy</requirement>
  </requirements>
  <tests>
    <!-- Test valid data -->
    <test>
      <param name="input1" value="1.bed"/>
      <param name="groupcol" value="1"/>
      <param name="ignorecase" value="true"/>
      <param name="optype" value="mean"/>
      <param name="opcol" value="2"/>
      <param name="opround" value="no"/>
      <output name="out_file1" file="groupby_out1.dat"/>
    </test>
    <!-- Long case but test framework doesn't allow yet
    <test>
      <param name="input1" value="1.bed"/>
      <param name="groupcol" value="1"/>
      <param name="ignorecase" value="false"/>
      <param name="operations" value='[{"opcol": "2", "__index__": 0, "optype": "mean", "opround": "no"}, {"opcol": "2", "__index__": 1, "optype": "median", "opround": "no"}, {"opcol": "6", "__index__": 2, "optype": "mode", "opround": "no"}, {"opcol": "2", "__index__": 3, "optype": "max", "opround": "no"}, {"opcol": "2", "__index__": 4, "optype": "min", "opround": "no"}, {"opcol": "2", "__index__": 5, "optype": "sum", "opround": "no"}, {"opcol": "1", "__index__": 6, "optype": "length", "opround": "no"}, {"opcol": "1", "__index__": 7, "optype": "unique", "opround": "no"}, {"opcol": "1", "__index__": 8, "optype": "cat", "opround": "no"}, {"opcol": "6", "__index__": 9, "optype": "cat_uniq", "opround": "no"}, {"opcol": "2", "__index__": 10, "optype": "random", "opround": "no"}, {"opcol": "2", "__index__": 11, "optype": "std", "opround": "no"}]'/>
      <output name="out_file1" file="groupby_out3.tabular"/>
    </test>
    -->
    <!-- Test data with an invalid value in a column. Can't do it because test framework doesn't allow testing of errors
    <test>
      <param name="input1" value="1.tabular"/>
      <param name="groupcol" value="1"/>
      <param name="ignorecase" value="true"/>
      <param name="optype" value="mean"/>
      <param name="opcol" value="2"/>
      <param name="opround" value="no"/>
      <output name="out_file1" file="groupby_out2.dat"/>
    </test>
     -->
  </tests>
  <help>

.. class:: infomark

**TIP:** If your data is not TAB delimited, use *Text Manipulation-&gt;Convert*

-----

**Syntax**

This tool allows you to group the input dataset by a particular column and perform aggregate functions: Mean, Median, Mode, Sum, Max, Min, Count, Concatenate, and Randomly pick on any column(s).

The Concatenate function will take, for each group, each item in the specified column and build a comma delimited list. Concatenate Unique will do the same but will build a list of unique items with no repetition.

Count and Count Unique are equivalent to Concatenate and Concatenate Unique, but will only count the number of items and will return an integer.

- If multiple modes are present, all are reported.

-----

**Example**

- For the following input::

   chr22  1000  1003  TTT
   chr22  2000  2003  aaa
   chr10  2200  2203  TTT
   chr10  1200  1203  ttt
   chr22  1600  1603  AAA

- **Grouping on column 4** while ignoring case, and performing operation **Count on column 1** will return::

   AAA    2
   TTT    3
   
- **Grouping on column 4** while not ignoring case, and performing operation **Count on column 1** will return::

   aaa    1
   AAA    1
   ttt    1
   TTT    2
  </help>
</tool>