biplot_color: Function for biplotting with no point labels and with points...

In YosefLab/scone: Single Cell Overview of Normalized Expression data

Function for biplotting with no point labels and with points color-coded according to a quantitative variable. For example: the rank of normalization performance.

Description

This function implements biplot for prcomp objects.

Usage

biplot_color(
  x,
  y,
  rank = TRUE,
  ties_method = c("max", "min", "first", "last", "random"),
  choices = 1:2,
  expand = 1,
  ...
)

Arguments

x	prcomp object.
y	numeric. Quantitative values used to color the points. If rank is FALSE, all values must be positive integers and less than or equal to the length of y.
rank	logical. If TRUE (default) y will be transformed by the rank() function
ties_method	character. ties.method used by the rank() function
choices	numeric. 2 principal components to plot. Default to first two PCs.
expand	numeric. value used to adjust the spread of the arrows relative to the points.
...	arguments passed to plot.

Value

Invisibly returns scaled point coordinates used in plot.

Examples

mat <- matrix(rnorm(1000), ncol=10)
colnames(mat) <- paste("X", 1:ncol(mat), sep="")

pc <- prcomp(mat)

biplot_color(pc, rank(pc$x[,1]))

YosefLab/scone documentation built on Oct. 21, 2024, 4:39 p.m.