tidyMicro (version 1.43)

micro_pca: Calculate and plot principle components

Description

Principle components are calculated on the centerted log ratio tranformation of the OTU table using the prcomp function from the stats package. Scaling the OTU table to a unit variance is the default option, and recommended, but this can be changed using scaled = F.

Usage

micro_pca(
  micro_set,
  table = NULL,
  dist = NULL,
  grp_var,
  y = clr,
  scale = TRUE,
  axes_arrows = F,
  main = NULL,
  subtitle = NULL,
  legend_title = NULL
)

Arguments

micro_set

A tidy_micro data set

table

OTU table of interest

dist

A distance matrix, such as a beta diversity. If supplied a PCoA plot will be returned

grp_var

Categorical grouping variable for color

y

Value to calculate principle components or coordinates on. Default is centered log ratio (recommended)

scale

Logical. Indicating whether the variables should be scaled to have unit variance before the analysis takes place

axes_arrows

Logical. Plot component axes arrows

main

Plot title

subtitle

Plot subtitle

legend_title

Legend title

Value

A ggplot you can add geoms to if you'd like

Details

PCA calculation is done by a singular value decomposition of the (centered and possibly scaled) data matrix, not by using eigen on the covariance matrix. This is generally the preferred method for numerical accuracy. Calculations are accomplished through the prcomp function, and the plot is created through internal code based on the ggbiplot function https://github.com/vqv/ggbiplot.

References

Becker, R. A., Chambers, J. M. and Wilks, A. R. (1988) The New S Language. Wadsworth & Brooks/Cole.

Mardia, K. V., J. T. Kent, and J. M. Bibby (1979) Multivariate Analysis, London: Academic Press.

Venables, W. N. and B. D. Ripley (2002) Modern Applied Statistics with S, Springer-Verlag.

Vincent Q. Vu (2011). ggbiplot: A ggplot2 based biplot. https://github.com/vqv/ggbiplot

Examples

# NOT RUN {
data(phy); data(cla); data(ord); data(fam); data(clin)

otu_tabs <- list(Phylum = phy, Class = cla, Order = ord, Family = fam)
set <- tidy_micro(otu_tabs = otu_tabs, clinical = clin) %>%
filter(day == 7) ## Only including first week

## PCA Plot
set %>% micro_pca(table = "Family", grp_var = bpd1)

## PCoA Plot (Recommended for p > n)

bray_beta <- set %>% beta_div(table = "Family")

set %>%
micro_pca(dist = bray_beta, grp_var = bpd1)

# }