This project is not currently active. We recommend referring to GustaME, which provides useful documentation of many of the methods we were exploring here.

microbiogeo

Repository for code, data, and analysis results of the biogeographical statistical methods comparison project (aka microbiogeo).

To run the code in this repository, you will need the following dependencies installed (versions tested against are in parentheses):

• QIIME base install (1.8.0)
• IPython (1.2.1)
• pyzmq (14.0.1)
• nose (1.3.0)
• R (3.0.2)
• R optparse package (1.0.2)
• R vegan package (2.0-10)
• R ape package (3.0-11)

The easiest way to install the Python dependencies is via pip, e.g.:

pip install numpy
pip install qiime
pip install ipython
pip install pyzmq
pip install nose

You will also need microbiogeo/code added to your PYTHONPATH and microbiogeo/code/scripts added to your PATH.

To run the unit tests, assuming you are in the microbiogeo directory:

nosetests code

To run the actual workflows, you will need to cd into the microbiogeo/code directory and start an IPython cluster with the number of cores/processors you'd like parallel jobs to be executed on. For example, the following command will start 4 IPython Engines:

ipcluster start --n=4