proportionality: Proportionality Distance

In abodein/timeOmics: Time-Course Multi-Omics data integration

Proportionality Distance

Description

proportionality is a wrapper that compute proportionality distance for a clustering result (pca, spca, pls, spls, block.pls, block.spls). and it performs a u-test to compare the median within a cluster to the median of the entire background set.

Usage

proportionality(X)

Arguments

X	an object of the class: pca, spca, pls, spls, block.pls or block.spls

Value

Return a list containing the following components:

propr.distance	Square matrix with proportionality distance between pairs of features
propr.distance.w.cluster	distance between pairs with cluster label
pvalue	Wilcoxon U-test p-value comparing the medians within clusters and with the entire background set

References

Lovell, D., Pawlowsky-Glahn, V., Egozcue, J. J., Marguerat, S., Bähler, J. (2015). Proportionality: a valid alternative to correlation for relative data. PLoS Comput. Biol. 11, e1004075. doi: 10.1371/journal.pcbi.1004075

Quinn, T. P., Richardson, M. F., Lovell, D., Crowley, T. M. (2017). propr: an r-package for identifying proportionally abundant features using compositional data analysis. Sci. Rep. 7, 16252. doi: 10.1038/s41598-017-16520-0

Examples

demo <- suppressWarnings(get_demo_cluster())

# pca
X <- demo$pca
propr.res <- proportionality(X)
plot(propr.res)

# pls
X <- demo$spls
propr.res <- proportionality(X)
plot(propr.res)

# block.pls
X <- demo$block.spls
propr.res <- proportionality(X)
plot(propr.res)

abodein/timeOmics documentation built on April 12, 2024, 12:30 p.m.