scikit-bio: a community-driven Python library for bioinformatics, providing versatile data structures, algorithms and educational resources.
Simplex space operations for compositional data implemented in Python.
Deep learning for personalized interpretability for compositional health data
Detection-limit censoring encodes survey identity and inflates reported cross-region generalization in pooled continental geochemical surveys. Full reproducible pipeline: censoring audit, dose-response simulation, mask experiments, four handling regimes, three validation protocols. Code and results for the Journal of Earth Science manuscript.
Recover the true composition of environmental samples from amplicon read counts distorted by per-taxon PCR efficiency. A hard, programmatically verified agent-evaluation task.
A multivariable MbWAS pipeline identifying taxonomic biomarkers of gut dysbiosis in Crohn's Disease. Implements zero-replaced Centered Log-Ratio (CLR) transformations and two-part hurdle regression models integrated via the Cauchy combination test to contrast the HMP2 / IBDMDB cohort.
Ternary compositional pipeline that quantifies whether tumors selectively co-opt extraembryonic (trophoblast) programs rather than embryonic-proper pluripotency, and identifies the master TFs driving each direction.
To associate your repository with the compositional-data topic, visit your repo's landing page and select "manage topics."