Using conda

Install Miniconda with Python 3. If you already have Miniconda installed with Python 2, download the latest Python 3 version and follow conda’s installation instructions. If you already have an older Miniconda version installed with Python 3, you may need to update your installation prior to installing Nextstrain’s tools with:

conda activate base
conda update conda

Create a conda environment to install augur into and activate that environment.

conda create -n nextstrain
conda activate nextstrain

Install augur and its dependencies into your environment.

conda install -c conda-forge -c bioconda augur

For a much faster installation process, use mamba as a drop-in replacement for conda.

conda install -c conda-forge mamba
mamba install -c conda-forge -c bioconda augur

Using pip from PyPi

Augur is written in Python 3 and requires at least Python 3.7. It’s published on PyPi as nextstrain-augur, so you can install it with pip like so:

python3 -m pip install nextstrain-augur

Augur uses some common external bioinformatics programs which you’ll need to install to have a fully functioning toolkit:

  • Nextstrain workflows and some tutorials require Snakemake

  • augur align requires mafft

  • augur tree requires at least one of:

  • Bacterial data (or any VCF usage) requires vcftools

On macOS, you can install most of these external programs using Homebrew with:

brew tap brewsci/bio
brew install mafft iqtree raxml fasttree vcftools

On Debian/Ubuntu, you can install them via:

sudo apt install mafft iqtree raxml fasttree vcftools

Other Linux distributions will likely have the same packages available, although the names may differ slightly. Follow Snakemake’s installation instructions for your operating system.

Install from source

git clone
python3 -m pip install .

This install depends on a fairly minimal set of external Python libraries. There are some functions in augur that require a larger set of dependencies. These can be installed via:

python3 -m pip install '.[full]'

If you wish to also install the development dependencies, and install augur in an “editable” mode whereby changes to the source code are reflected in your version of augur then run:

python3 -m pip install -e '.[dev]'

See above for how to install the external bioinformatics programs which you’ll need to have a fully functioning toolkit.

Testing if it worked

If installation worked, you should be able to run augur --help and see augur’s primary help output.