View source: R/plot_landscape.R
plot_landscape | R Documentation |
Wrapper for visualizing sample similarity landscape ie. sample density in various 2D projections.
plot_landscape(
x,
method = "PCoA",
distance = "bray",
transformation = "identity",
col = NULL,
main = NULL,
x.ticks = 10,
rounding = 0,
add.points = TRUE,
adjust = 1,
size = 1,
legend = FALSE,
shading = TRUE,
shading.low = "#ebf4f5",
shading.high = "#e9b7ce",
point.opacity = 0.75
)
x |
|
method |
Ordination method, see phyloseq::plot_ordination; or "PCA", or "t-SNE" (from the Rtsne package) |
distance |
Ordination distance, see phyloseq::plot_ordination; for method = "PCA", only euclidean distance is implemented now. |
transformation |
Transformation applied on the input object x |
col |
Variable name to highlight samples (points) with colors |
main |
title text |
x.ticks |
Number of ticks on the X axis |
rounding |
Rounding for X axis tick values |
add.points |
Plot the data points as well |
adjust |
Kernel width adjustment |
size |
point size |
legend |
plot legend TRUE/FALSE |
shading |
Add shading in the background. |
shading.low |
Color for shading low density regions |
shading.high |
Color for shading high density regions |
point.opacity |
Transparency-level for points |
For consistent results, set random seet (set.seed) before function call. Note that the distance and transformation arguments may have a drastic effect on the outputs.
A ggplot
plot object.
data(dietswap)
# PCoA
p <- plot_landscape(transform(dietswap, "compositional"),
distance = "bray", method = "PCoA")
p <- plot_landscape(dietswap, method = "t-SNE", distance = "bray",
transformation = "compositional")
# PCA
p <- plot_landscape(dietswap, method = "PCA", transformation = "clr")
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.