clustNet: Network-based clustering with covariate adjustment

License: GPL v3

clustNet is an R package for network-based clustering of categorical data using a Bayesian network mixture model and optional covariate adjustment.


The package requires Rgraphviz and RBGL, which can be installed from Bioconductor as follows:

{r eval=FALSE} if (!requireNamespace("BiocManager", quietly = TRUE)) install.packages("BiocManager") BiocManager::install(c("Rgraphviz", "RBGL"))

The latest stable version of clustNet is available on CRAN and can be installed with

{r eval=FALSE} install.packages("clustNet") from within an R session. On a normal computer, this should take around 5-60 seconds, depending on how many of the required packages are already installed.


Being hosted on GitHub, it is also possible to use the install_github tool from an R session to install the latest development version:

{r eval=FALSE} library("devtools") install_github("cbg-ethz/clustNet")

clustNet requires R >= 3.5.


```{r eval=FALSE} library(clustNet)

Simulate data

k_clust <- 3 # numer of clusters ss <- c(400, 500, 600) # samples in each cluster simulation_data <- sampleData(k_clust = k_clust, n_vars = 20, n_samples = ss) sampled_data <- simulation_data$sampled_data

Network-based clustering

cluster_results <- get_clusters(sampled_data, k_clust = k_clust)

Load additional pacakges to visualize the networks

library(ggplot2) library(ggraph) library(igraph) library(ggpubr)

Visualize networks


Load additional pacakges to create a 2d dimensionality reduction

library(car) library(ks) library(graphics) library(stats)

Plot a 2d dimensionality reduction



On a normal computer, the clustering should take around 2-4 minutes.