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.
-
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).
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.
The new tutorial demonstrates the new estimation and calibration functions on a new dataset from Leopold and Busby (2020).
- Fixed failure of
coocurrence_matrix()when names were missing (#5).
Direct implementation of the methods described in McLaren MR, Willis AD, Callahan BJ (2019) and used for the analysis of that paper.