New features
New estimation and calibration functions
estimate_bias() and calibrate() provide easy-to-use high-level interfaces to the original metacal bias-estimation and calibration method. Their use is illustrated in the new tutorial.
New utility functions
-
pairwise_ratios()allows computing ratios between pairs of taxa and/or samples. -
perturb()applies a compositional perturbation to all observations in an abundance matrix (or phyloseq object).
Support for phyloseq objects
The new functions estimate_bias(), perturb(), calibrate(), and pairwise_ratios() all work with objects from the phyloseq package as well as plain matrices. Phyloseq is now a required dependency, though may be made optional in the future.
New tutorial
The new tutorial demonstrates the new estimation and calibration functions on a new dataset from Leopold and Busby (2020).
Minor fixes
- Fixed failure of
coocurrence_matrix()when names were missing (#5).