Case study: Statistical testing via PROMPT's Java API

Keywords:

Mann-Whitney, Kolmogorov Smirnov, Fisher's exact test, Chiq-Square test, scripting, Java API, direct call

Prerequistes:

For compilation and executation you'll need either the prompt_complete.jar or the prompt_slim.jar file in your classpath from the download page. We recommend to use the prompt_complete.jar as this big file contains all jars that may be needed whereas the slim version just contains the prompt-binaries and further tools like logger or xml access libraries must be added manually. Optionally you may have a look at the source code and/or the PROMPT JavaDoc.

All test classes need R or S-Plus as statistical calculation engine. If the R or S executables or linking libraries are not found automatically you may need to specific the location manually. More information can be found at the help page, section How to set the R-Path.

Questions 1: Are two numeric sets differentially distributed?

Examples are in perparation

Question 2: Is one or more categories over- or underrepresented?

The following two Java programs are testing wether the differentially expressed (DE) genes are significantly enriched or depleted in the GO category/categories of interest. As input the continceny table is used directly as introduced at the overview page is used.

Demo1.java
Reads the contincency table from a tab delimited text file data.txt.

Demo2.java
Creates a data object that holds the contincency table and passes the data object directly to the test class. This has the advantage as no temporary file needs to be created.