galaxy/tools/regVariation/t_test_two_samples.xml

<tool id="t_test_two_samples" name="T Test for Two Samples" version="1.0.0">
  <description></description>
  
  <command interpreter="perl">
  	t_test_two_samples.pl $inputFile1 $inputFile2 $inputTestSidedness3 $inputStandardDeviationEquality4 $outputFile1
  </command>

  <inputs>
  	<param format="tabular" name="inputFile1" type="data" label="Select the first sample tabular file"/>
  	<param format="tabular" name="inputFile2" type="data" label="Select the second sample tabular file"/>
  	
    <param name="inputTestSidedness3" type="select" label="Choose the test sidedness:">
    	<option value="two-sided">Two-sided</option>
      	<option value="one-sided:_m1_less_than_m2">One-sided: m1 less than m2</option>
      	<option value="one-sided:_m1_greater_than_m2">One-sided: m1 greater than m2</option>
    </param>
    
    <param name="inputStandardDeviationEquality4" type="select" label="Choose the standard deviation equality status of the two populations:">
    	<option value="equal">Equal</option>
      	<option value="unequal">Unequal</option>
    </param>
  </inputs>
  
  <outputs>
    <data format="text" name="outputFile1"/>
  </outputs>
  
  <tests>
  	<test>
  		<param name="inputFile1" value="sample1.tabular" ftype="tabular" />
  		<param name="inputFile2" value="sample2.tabular" ftype="tabular" />
    	<param name="inputTestSidedness3" value="Two-sided" />
    	<param name="inputStandardDeviationEquality4" value="Equal" />
    	<output name="outputFile1" file="t_test_result1.text" />
  	</test>
  	
  	<test>
  		<param name="inputFile1" value="sample1.tabular" ftype="tabular" />
  		<param name="inputFile2" value="sample2.tabular" ftype="tabular" />
    	<param name="inputTestSidedness3" value="Two-sided" />
    	<param name="inputStandardDeviationEquality4" value="Unequal" />
    	<output name="outputFile1" file="t_test_result2.text" />
  	</test>
  	
  	<test>
  		<param name="inputFile1" value="sample1.tabular" ftype="tabular" />
  		<param name="inputFile2" value="sample2.tabular" ftype="tabular" />
    	<param name="inputTestSidedness3" value="One-sided: m1 less than m2" />
    	<param name="inputStandardDeviationEquality4" value="Equal" />
    	<output name="outputFile1" file="t_test_result3.text" />
  	</test>
  	
  	<test>
  		<param name="inputFile1" value="sample1.tabular" ftype="tabular" />
  		<param name="inputFile2" value="sample2.tabular" ftype="tabular" />
    	<param name="inputTestSidedness3" value="One-sided: m1 less than m2" />
    	<param name="inputStandardDeviationEquality4" value="Unequal" />
    	<output name="outputFile1" file="t_test_result4.text" />
  	</test>
  	
  	<test>
  		<param name="inputFile1" value="sample1.tabular" ftype="tabular" />
  		<param name="inputFile2" value="sample2.tabular" ftype="tabular"/>
    	<param name="inputTestSidedness3" value="One-sided: m1 greater than m2" />
    	<param name="inputStandardDeviationEquality4" value="Equal" />
    	<output name="outputFile1" file="t_test_result5.text" />
  	</test>
  	
  	<test>
  		<param name="inputFile1" value="sample1.tabular" ftype="tabular" />
  		<param name="inputFile2" value="sample2.tabular" ftype="tabular" />
    	<param name="inputTestSidedness3" value="One-sided: m1 greater than m2" />
    	<param name="inputStandardDeviationEquality4" value="Unequal" />
    	<output name="outputFile1" file="t_test_result6.text" />
  	</test>
  </tests>


  <help> 

.. class:: infomark

**What it does**

This program implements the non-pooled t-test for two samples where the alternative hypothesis is two-sided or one-sided. The program takes four inputs:

- The first input file is a TABULAR format file representing the first sample and consisting of one column only.
- The second input file is a TABULAR format file representing the first sample and consisting of one column only.
- The third input is the sidedness of the t-test: either two-sided or, one-sided with m1 less than m2 or, one-sided with m1 greater than m2. 
- The fourth input is the equality status of the standard deviations of both populations.
- The output file is a TXT file representing the result of the two-sample t-test.


**Example**

Let us have the first input file representing the first sample as follows::

	5
	4
	8
	6
	7
	2
	1
	1
	0
	6
	4
	5
	7
	5
	3
	2
	5
	8
	7
	6
	4

And the second input file representing the second sample as follows::

	2
	3
	5
	1
	2
	7
	5
	4
	3
	2
	7
	6
	0
	8
	4
	6
	9
	2
	4
	5
	6

Runnig the program and choosing "Two-sided" and "Equal" as parameters will give the following output::

	Two Sample t-test

	data:  sample1 and sample2 
	t = -0.3247, df = 40, p-value = 0.7471
	alternative hypothesis: true difference in means is not equal to 0 
	95 percent confidence interval:
 	-1.720030  1.243839 
	sample estimates:
	mean of x mean of y 
 	4.333333  4.571429 


  </help>  
  
</tool>