Scatter plot

Produces a plot describing the relationship between two columns of the outcomes matrix \(Y\). Colors and symbols can be chosen for the levels of the design factors. Ellipses, polygons or segments can be added to group different sets of points on the graph.

Usage

plotScatter(
  Y = NULL,
  design = NULL,
  lmpDataList = NULL,
  xy,
  color = NULL,
  shape = NULL,
  points_labs = NULL,
  title = "Scatter plot",
  xlab = NULL,
  ylab = NULL,
  size = 2,
  size_lab = 3,
  drawShapes = c("none", "ellipse", "polygon", "segment"),
  typeEl = c("norm", "t", "euclid"),
  levelEl = 0.9,
  alphaPoly = 0.4,
  theme = theme_bw(),
  drawOrigin = FALSE
)

Arguments

Y

A \(n \times m\) matrix with \(n\) observations and \(m\) variables. Can be NULL if lmpDataList is defined.

design

A \(n \times k\) "freely encoded" experimental design data.frame. Can be NULL if lmpDataList is defined.

lmpDataList

If not NULL, a list with outcomes, design and formula, as outputted by data2LmpDataList.

xy

x- and y-axis values: a vector of length 2 with either the column name(s) of the \(Y\) matrix to plot (character) or the index position(s) (integer).

color

If not NULL, a character string giving the column name of design to be used as color. Currently treated as a discrete variable.

shape

If not NULL, a character string giving the column name of design to be used as shape. Currently treated as a discrete variable.

points_labs

If not NULL, a character vector with point labels.

title

Plot title.

xlab

If not NULL, label for the x-axis.

ylab

If not NULL, label for the y-axis.

size

The points size, by default 2.

size_lab

The size of points labels, by default 3.

drawShapes

Multiple shapes can be drawn based on the color: "none" for no shape (default), "ellipse" (ellipses with ggplot2::stat_ellipse()), "polygon" (polygons with ggplot2::geom_polygon()) or "segment" (segment from the centroids with ggplot2::geom_segment()).

typeEl

The type of ellipse, either "norm" (multivariate normal distribution, the default), "t" (multivariate t-distribution) or "euclid" (draws a circle with the radius equal to level, representing the euclidean distance from the center).

levelEl

The confidence level at which to draw an ellipse, by default 0.9.

alphaPoly

The degree of transparency for polygons, by default 0.4.

theme

The ggplot2 theme (default: theme_bw()), see ?ggtheme for more info.

drawOrigin

If TRUE, draws horizontal and vertical intercepts at (0,0).

Value

A ggplot2 scatter plot.

Details

Either Y or lmpDataList need to be defined. If both are given, the priority goes to Y. The same rule applies for design or lmpDataList.

Examples

data("UCH")

# Without the design info
plotScatter(Y = UCH$outcomes, xy = c(453, 369))


# equivalent to:
plotScatter(lmpDataList = UCH, xy = c(453, 369))


# With color and shape
plotScatter(
  lmpDataList = UCH,
  xy = c(453, 369), color = "Hippurate",
  shape = "Citrate"
)


# equivalent to:
plotScatter(
  Y = UCH$outcomes, design = UCH$design,
  xy = c(453, 369), color = "Hippurate",
  shape = "Citrate"
)


# With color and shapes
plotScatter(
  Y = UCH$outcomes, design = UCH$design,
  xy = c(453, 369), color = "Hippurate",
  drawShapes = "ellipse"
)


plotScatter(
  Y = UCH$outcomes, design = UCH$design,
  xy = c(453, 369), color = "Hippurate",
  drawShapes = "polygon"
)


plotScatter(
  Y = UCH$outcomes, design = UCH$design,
  xy = c(453, 369), color = "Hippurate",
  drawShapes = "segment"
)


# With customized shapes
library(ggplot2)
plotScatter(
  Y = UCH$outcomes, design = UCH$design,
  xy = c(453, 369), shape = "Hippurate", size = 3
) +
  scale_discrete_identity(
    aesthetics = "shape",
    guide = "legend"
  )


plotScatter(
  Y = UCH$outcomes, design = UCH$design,
  xy = c(453, 369), shape = "Hippurate"
) +
  scale_shape_discrete(solid = FALSE)


plotScatter(
  Y = UCH$outcomes, design = UCH$design,
  xy = c(453, 369), shape = "Hippurate"
) +
  scale_shape_manual(values = c(15, 16, 17))


# With labels
plotScatter(
  Y = UCH$outcomes, design = UCH$design,
  xy = c(453, 369), points_labs = rownames(UCH$design)
)