Installing scikit-bio

Scikit-bio can be installed via PyPI, Conda, system package managers, from source, or from pre-compiled wheels. The PyPI and Conda methods are the most commonly used.

System requirements

Platform

Scikit-bio runs on Linux, macOS, and Windows. It is currently available for the x86_64 and ARM64 architectures, although ARM64 support is not yet available for the Windows operating system.

Python

scikit-bio requires Python 3.9 or later installed in your system. See the Python version support section for more details.

Environment based

Conda

The recommended way to install scikit-bio is via the Conda package manager. The latest release of scikit-bio is distributed via the conda-forge channel. You can install it via the following command:

conda install -c conda-forge scikit-bio

Other channels such as anaconda and bioconda also host scikit-bio, which however may or may not be the up-to-date version.

PyPI

Alternatively, the latest release of scikit-bio can be installed from PyPI:

pip install scikit-bio

System package managers

Scikit-bio is available as third-party packages from software repositories for multiple Linux/BSD distributions. However, these packages may or may not be the latest version. The scikit-bio development team is not involved in the maintenance of these packages.

For example, users of Debian-based Linux distributions (such as Ubuntu and Linux Mint) may install scikit-bio using:

sudo apt install python3-skbio python-skbio-doc

Users of Arch Linux or variants (such as Manjaro) may install scikit-bio from AUR:

yay -S python-scikit-bio

Pre-compiled wheels

Starting with version 0.7.0, scikit-bio now provides pre-compiled wheels for each release. To install from a wheel file, download the appropriate file for your platform and run:

pip install your_wheel_file.whl

Development version

Scikit-bio is undergoing expansion, with many new features being introduced. You are welcome to try these features by installing the current development version from our GitHub repo:

pip install git+https://github.com/scikit-bio/scikit-bio.git

Alternatively, you may download the repository, extract, and execute:

pip install .

However, be cautious that the new functionality may not be stable and could be changed in the next formal release. It is not recommended to deploy the development version in a production environment.

Verifying the installation

After installing scikit-bio, verify the installation by running the following in a Python shell or script:

import skbio
print(skbio.__version__)

This should print the installed version of scikit-bio without errors.

For a more robust verification of the installation you may run the scikit-bio’s unit tests in the environment where scikit-bio is installed. This will require the pytest package to be installed:

python -m skbio.test

Acceleration

The new scikit-bio-binaries package is a separate package from scikit-bio, written in C++, which when installed in the same environment as scikit-bio will dramatically increase performance of select functions. Installation of scikit-bio-binaries is currently available with conda:

conda install -c conda-forge scikit-bio-binaries

Python version support

Scikit-bio’s policy is to support Python versions until they reach official end-of-life, which typically provides 4-5 years of support per version. This approach accounts for the prevalence of legacy and under-maintained software packages in bioinformatics workflows, where users often cannot update their environments frequently. The goal is to strike an appropriate balance between encouraging modernization and maintaining compatibility for our user base, while still providing predictable deprecation timelines that researchers can plan around.