canCor analysis cor correlation to principal components — plot_precovid

Performs PCA and correlates PCs to selected metadata using variancePartition::canCorPairs.

Usage

plot_precovid_cca(
  data_to_pca,
  sample_metadata,
  to_scale = TRUE,
  custom_title = "",
  interesting_variables = c("pid", "randomGroupCode", "Timepoint", "Sex", "BMI",
    "calculatedAge"),
  num_pcs = 5
)

Arguments

data_to_pca: The actual numerical data that should be included in analysis. must be able to be converted into a numerical matrix
sample_metadata: The correponding metadata to data_to_pca. The rownames of this matrix must correspond to colnames of the data matrix.
to_scale: logical; Whether to scale the data when performing PCA.
custom_title: character; whatever title is desired for the figure
interesting_variables: character; a vector of variables that are going to be correlated using canCors to the PCs. The # of PCs in num_pcs are included by default.
num_pcs: numeric; the number of principal components desired to be included.

Value

a pheatmap object with the desired correlations to principal components

Author

Christopher Jin