panel.cor.pval

<p>number of non-zero digits to leave at the end</p>

digits

cex.cor

p.cut

<p>works as upper.panel or lower.panel argument of pairs() (package graphics). Displays pvalues better than 0.1.</p>

Poisson-lognormal generalized linear mixed model analysis of multivariate counts data using MCMC, aiming to infer the changes in relative proportions of individual variables. The package was originally designed for sequence-based analysis of microbial communities ("metabarcoding", variables = operational taxonomic units, OTUs), but can be used for other types of multivariate counts, such as in ecological applications (variables = species). The results are summarized and plotted using 'ggplot2' functions. Includes functions to remove sample and variable outliers and reformat counts into normalized log-transformed values for correlation and principal component/coordinate analysis. Walkthrough and examples: http://www.bio.utexas.edu/research/matz_lab/matzlab/Methods_files/walkthroughExample_mcmcOTU_R.txt.

panel.cor.pval: accessory function for pairs() to display pvalue of the Pearson correlation

Description

Usage

Arguments

References