K-Pax2 has been superseded by Kpax3.jl, a package for the Julia programming language (that can be used with R through RJulia). Kpax3.jl provides the same functionalities of K-Pax2 but in a faster and better way. It also add a MCMC algorithm for a proper Bayesian analysis. Check it out!
K-Pax2 is a R package, written with the purpose of clustering (big) datasets of categorical statistical variables. Main application of K-Pax2 is with genetic datasets, such as dna/protein multiple sequence alignments. Being a general method, it can be easily applied to any kind of categorical dataset. K-Pax2 output consists of a classification of both the rows (statistical units) and columns (statistical variables) of the provided data matrix
To know more about the method and how to properly setup the parameters, please read the following paper:
Pessia A., Grad Y., Cobey S., Puranen J., Corander J. (2015). K-Pax2: Bayesian identification of cluster-defining amino acid positions in large sequence datasets. MGen 1(1). doi: 10.1099/mgen.0.000025
Visit also the BSG group website for other interesting projects.
The easiest way to install K-Pax2 is using devtools:
devtools::install_github("albertopessia/kpax2/kpax2")
library(kpax2)
Download the latest version of K-Pax2 from the releases page and extract the content of the compressed file into a folder of your choice. Let "PathToPackageFile" denote the complete path to the package file (having extension .tar.gz).
Decide if the package should be made available only to the current user or to every user on the same machine. In the latter case, run all the following commands with administrative privileges.
Note for Windows users: within R, file paths should be written using the slash '/' character instead of the usual backslash '' character. If you want to use the backslash, you have to escape it. Examples of valid file paths:
"C:/Users/CurrentUser/Documents/kpax2/kpax2_x.x.x.tar.gz"
"C:\\Users\\CurrentUser\\Documents\\kpax2\\kpax2_x.x.x.tar.gz"
If you want to install the package into the default system library, run R and type the command
install.packages("PathToPackageFile", repos=NULL, type="source")
An alternative approach is to open the terminal and issue the command
R CMD INSTALL "PathToPackageFile"
If it fails to find R, then it means that it is not in the system path. Ask your machine administrator for help.
You can now load the package in R using
library(kpax2)
If you want to install the package into a custom library, different from the default system one, choose its location and denote with "PathToLibrary" its complete path. Open the terminal and issue the command
R CMD INSTALL --library="PathToLibrary" "PathToPackageFile"
You can now load the package in R using
library(kpax2, lib.loc="PathToLibrary")
For a tutorial on how to properly use this package, follow the instructions written in the file kpax2_tutorial.R.
See LICENSE.md