MATSS-LDATS

Contributors

Overview

This repository builds on analytical methods and infrastructure in development elsewhere in our lab group to perform a meta-analysis on the dynamics of community structure in ecological timeseries. We use the data and pipeline management tools in MATSS to collect ecological time series data, and perform the LDA and timeseries analyses in LDATS. We are especially curious about the usual number of topics and changepoints detected in community timeseries, and whether these characteristics are related to features of the data (timeseries length, sampling frequency) or the system in question (study taxa, species richness, abundance).

To use

Setup

install MATSS

Because MATSS is in active and rapid development, we suggest re-installing MATSS whenever you pick up this analysis.

The same goes for drake.

System/environmental variables

For storing datasets downloaded with the retriever. This only needs to be done once per machine.

MATSS and drake

Related repos & reading

https://github.com/weecology/MATSS

https://github.com/weecology/LDATS

https://github.com/emchristensen/Extreme-events-LDA

To use on HiperGator

Follow the instructions in the MATSS repo: https://weecology.github.io/MATSS/articles/hipergator-install.html

In R on HiperGator, install MATSS-LDATS:

remotes::install_github('weecology/MATSS-LDATS')

Then in shell, navigate to our group's shared folder:

cd orange/ewhite/MATSS-LDATS

Run pipeline.R:

Rscript analysis/pipeline.R