clustree's Introduction

clustree

Deciding what resolution to use can be a difficult question when approaching a clustering analysis. One way to approach this problem is to look at how samples move as the number of clusters increases. This package allows you to produce clustering trees, a visualisation for interrogating clusterings as resolution increases.

Installation

clustree is currently unavailable from CRAN, please use the GitHub installation instructions

You can install the release version of clustree from CRAN with:

install.packages("clustree")

If you want to use the development version that can be installed from GitHub using with:

# install.packages("devtools")
devtools::install_github("lazappi/clustree")

To also build the vignettes use:

# install.packages("devtools")
devtools::install_github("lazappi/clustree", dependencies = TRUE,
                         build_vignettes = TRUE)

NOTE: Building the vignettes requires the installation of additional packages.

Documentation

The documentation for clustree is available from CRAN at https://cran.r-project.org/package=clustree.

To view the vignette and all the package documentation for the development version visit http://lazappi.github.io/clustree.

Citing clustree

If you use clustree in your work please cite our preprint "Zappia L, Oshlack A. Clustering trees: a visualisation for evaluating clusterings at multiple resolutions. bioRxiv. 2018; doi:10.1101/274035".

  @Article{,
    author = {Luke Zappia and and Alicia Oshlack},
    title = {Clustering trees: a visualisation for evaluating clusterings at
              multiple resolutions},
    journal = {bioRxiv},
    year = {2018},
    url = {https://doi.org/10.1101/274035},
    doi = {10.1101/274035},
  }