eigenVariable: Association of eigen vectors with variables
In moosik/snorm: Supporting function for supervised exploration of data structure

Description Usage Arguments Value See Also Examples

This function calculates P values for association of any number of eigen vectors with the metadata variables. Method: anova(lm(eigen vector ~ variable))$"Pr(>F)"[1]

1	eigenVariable(pcs, tech.vars)

`pcs`	a matrix with eigen vectors, i.e. V matrix of an SVD
`tech.vars`	data frame with the metadata variables

result: Matrix with the P values where columns are the eigen vectors and the rows are the variables from the metadata
plot: ggplot object for a tile plot reflecting significant associations

anova

u <- matrix(runif(120), ncol = 3)
library(ggplot2)
var.test <- data.frame(batch = sample(c("a", "b"), 40, replace = TRUE),
                       age = sample(50:85, size = 40, replace = TRUE),
                       amp.plate = sample(c("A1", "B1", "C1", "D1", "E1"), 40, replace = TRUE),
                       height = sample(60:72, 40, replace = TRUE))
res <- eigenVariable(u, var.test)
print(res$result)
plot(res$plot)