A QIIME 2 plugin for solving constrained sparse regression and classification problems with microbiome data including:
- Sparse log-contrast regression
- Cross-validation for hyperparameter selection
- Stability selection for feature selection
- Classification and regression tasks
- Tree-aggregated predictive modeling (trac)
- Interactive visualizations with model diagnostics
📂 Tutorial & Examples
Installation¶
First, make sure you have the required dependencies installed:
conda install -c conda-forge zarr plotly
OR
pip install zarr plotly c-lasso
Next to install the plugin:
pip install git+https://github.com/bio-datascience/q2-classo-latest.git
qiime dev refresh-cache
Usage Tutorial¶
A complete tutorial on using q2-classo
for microbiome data analysis — including preprocessing, CLR transformation, constrained regression, and visualization — is available in the repository examples.
👉 Quick Start Guide
This tutorial includes:
- Random data generation and basic workflow
- CLR transformation and taxonomic aggregation
- Constrained lasso regression with cross-validation
- Stability selection for robust feature selection
- HIV sCD14 prediction case study
- Interactive visualization of model results
Citation¶
If you use q2-classo
, please cite:
Bien, J., Yan, X., Simpson, L. and Müller, C. L. (2020). Tree-Aggregated Predictive Modeling of Microbiome Data. arXiv preprint arXiv:2002.08698.
Related Projects¶
c-lasso
: Python solvers for constrained lasso problemsq2-gglasso
: QIIME 2 plugin for graphical lasso problems- QIIME 2: Extensible microbiome analysis platform
License¶
BSD 3-Clause License. See LICENSE for details.
- Links
- Documentation
- Source Code
- Stars
- 0
- Last Commit
- f7056e0
- Available Distros
- 2025.4
- 2025.4/amplicon