NetGAM

By: Samantha J. Gleich, Jacob A. Cram, Jake L. Weissman, and David A. Caron

Last Modified: October 17, 2021

NetGAM is an R package that is designed to decrease the influence that time-series properties (i.e. seasonality, long-term trends, and autocorrelation) have on ecological network analyses. The NetGAM package takes species abundance data as input and uses generalized additive mixed models (gamms) in the mgcv package to model each species as a function of time-series predictor variables. The residuals of each gamm are then extracted and used in downstream networking methods (e.g. graphical lasso, correlation networks, etc.).

The NetGAM package requires the use of other R packages:

• mgcv
• dplyr
• compositions
• stats
• psych
• pulsar
• batchtools
• huge

Installation

The NetGAM package can be installed using the devtools package

install.packages("devtools")
library(devtools)
devtools::install_github("sgleich/NetGAM")
library(NetGAM)

The NetGAM package dependencies can also be installed as follows:

install.packages(c("mgcv","dplyr","compositions","stats","psych","pulsar","batchtools","huge"))

Example

An example of how to use the NetGAM package is outlined here