Principal Component Analysis (for AMR)

Performs a principal component analysis (PCA) based on a data set with automatic determination for afterwards plotting the groups and labels.

# S3 method for data.frame
prcomp(
  x,
  ...,
  retx = TRUE,
  center = TRUE,
  scale. = TRUE,
  tol = NULL,
  rank. = NULL
)

pca(x, ...)

Arguments

x	a data.frame containing numeric columns
...	columns of x to be selected for PCA
retx	a logical value indicating whether the rotated variables should be returned.
center	a logical value indicating whether the variables should be shifted to be zero centered. Alternately, a vector of length equal the number of columns of x can be supplied. The value is passed to scale.
scale.	a logical value indicating whether the variables should be scaled to have unit variance before the analysis takes place. The default is FALSE for consistency with S, but in general scaling is advisable. Alternatively, a vector of length equal the number of columns of x can be supplied. The value is passed to scale.
tol	a value indicating the magnitude below which components should be omitted. (Components are omitted if their standard deviations are less than or equal to tol times the standard deviation of the first component.) With the default null setting, no components are omitted (unless rank. is specified less than min(dim(x)).). Other settings for tol could be tol = 0 or tol = sqrt(.Machine$double.eps), which would omit essentially constant components.
rank.	optionally, a number specifying the maximal rank, i.e., maximal number of principal components to be used. Can be set as alternative or in addition to tol, useful notably when the desired rank is considerably smaller than the dimensions of the matrix.

Details

The pca() function takes a data.frame as input and performs the actual PCA with the R function prcomp().

The result of the pca() function is a prcomp object, with an additional attribute non_numeric_cols which is a vector with the column names of all columns that do not contain numeric values. These are probably the groups and labels, and will be used by ggplot_pca().

Experimental lifecycle

The lifecycle of this function is experimental. An experimental function is in the very early stages of development. The unlying code might be changing frequently as we rapidly iterate and explore variations in search of the best fit. Experimental functions might be removed without deprecation, so you are generally best off waiting until a function is more mature before you use it in production code. Experimental functions will not be included in releases we submit to CRAN, since they have not yet matured enough.

Examples

# `example_isolates` is a dataset available in the AMR package.
# See ?example_isolates.

# calculate the resistance per group first
library(dplyr)
resistance_data <- example_isolates %>%
  group_by(order = mo_order(mo),       # group on anything, like order
           genus = mo_genus(mo)) %>%   #  and genus as we do here
  summarise_if(is.rsi, resistance)     # then get resistance of all drugs

# now conduct PCA for certain antimicrobial agents
pca_result <- resistance_data %>%
  pca(AMC, CXM, CTX, CAZ, GEN, TOB, TMP, SXT)

pca_result
summary(pca_result)
biplot(pca_result)
ggplot_pca(pca_result) # a new and convenient plot function