Skip to content

stenver/wilcoxon-test

Repository files navigation

#Wilcoxon Signed-rank test

This project is an implementation of Wilcoxon signed-rank test.

For more information about the test, read: http://en.wikipedia.org/wiki/Wilcoxon_signed-rank_test

The goal of this project is to create a C++ library that can be run from R statistics program and terminal. The library should be able to run tens of thousands of Wilcoxon Signed Ranked Tests in in mere seconds. It should also be able to run these tests relatevly accurately, regardless of the sample size(The number of pairs) and take into account that some test might be missing or flawed. The library is developed for BIIT(Bioinformatics, Algorithmics and Data mining group). BIIT is joint research group between the Department of Computer Science (University of Tartu), Quretec, and the Estonian Biocenter. Its main research topics and capabilities include the gene regulation, gene expression data analysis, biological data mining and others.

Note that you must use Linux system to use this repository and it has been tested on Linux Mint KDE and Ubuntu distributions.

The repository contains a number of folders.

##WilcoxonTestLibrary Implementation of the wilcoxon test shared library. Since this is a shared library, not an executable, you cannot run it. However, you can install (and create new) interfaces for it, some of which are available in this project. The library can be installed by going to the folder and running:

>make install

After installing, you can run tests of the project with the command:

>make test
>./runTests

To get verbose output, use the command:

>./runTests -v

##RcppWilcoxonTest An interface that connects the implementation of the optimized Wilcoxon algorithm to the R. Note that you must have installed the implementation to use it.

You need Rcpp packages for R to use it.

To install Rcpp, open R console and run:

$install.packages("Rcpp")

After that, the R wilcoxon package can be installed by navigating to the package folder in the terminal and running the command:

$R CMD INSTALL .

You can now load the library in R by calling:

>library('RcppWilcoxonTest')

you can invoke the function by calling:

>RcppWilcoxonTest::WilxTest(dataMatrix, dataXsize, dataYsize, testIndexes, controlIndexes)

##TerminalWilcoxonTest An interface that connects the implementation Wilcoxon library to the terminal.

Note that you must have installed the implementation to use it.

The interface can be compiled by going to the folder and running

$make

You can then run

$WilcoxonTest --help

for further help. Currently only supports NetCDF file as input data.

##WilcoxonVTable Python program that can calculate V and P tables, print them, create files of the tables and create a number of graphs on the tables. These methods were mostly made for caulculations and research and are not structured very well, nor meant to be used publicly. Use them at your own risk.

If you want to run tests, use the following command in the test folder

python3 -m unittest discover . '*Test.py'

##Seminar_paper The folder that contains this paper and all images attached to it. It also contains R programs to create those images.

For any suggestions, send me a message.

#Troubleshooting If R cant run it - make sure that you have the latest version of R and Rcpp installed.

In case of any errors, contact me directly.

About

No description, website, or topics provided.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages