EMtree

EMtree infers direct species association networks, implementing the procedure described in Momal et al.. This package uses averages over spanning trees within an Expectation-Maximization algorithm to infer conditional dependence networks, and involves plotting functionalities (using ggraph and tydigraph).

By default, it uses the Poisson log-Normal Model (PLNmodels) to accommodate abundance data. However, EMtree is an inference method which only requires an estimate of a Gaussian covariance matrix, and can be used with any model which either use Gaussian latent variables, Gaussian copulas, or Gaussian data transformations.

Installation

EMtree requires R>3.5.

CRAN dependencies

required_CRAN <- c("Matrix", "purrr","parallel",  "mvtnorm", "vegan","huge",
                   "ggplot2", "magrittr", "dplyr","tidyr", "tibble",
                   "PLNmodels","ggraph", "tidygraph", "ade4")
not_installed_CRAN <- setdiff(required_CRAN, rownames(installed.packages()))
if (length(not_installed_CRAN) > 0) install.packages(not_installed_CRAN)

Installation of EMtree

You can install the development version of EMtree:

devtools::install_github("Rmomal/EMtree")

Reference

Please cite our work using the following reference:

  • Momal, R., Robin, S., & Ambroise, C. (2020). Tree‐based inference of species interaction networks from abundance data. Methods in Ecology and Evolution, 11(5), 621-632.