Computes t-SNE into two dimensions and plots the map as hexagonal bins,
where the value of the coloring variable is summarised for each bin, by
default as the mean of the values inside the bin.
In case there are missing values, PCA is performed using the nipals method
of pca, the method can be changed to "ppca" if
nipals fails.
Usage
plot_tsne_hexbin(
object,
all_features = FALSE,
center = TRUE,
scale = "uv",
pca_method = "nipals",
perplexity = 30,
fill = "Injection_order",
summary_fun = "mean",
bins = 10,
title = "t-SNE",
subtitle = paste("Perplexity:", perplexity),
fill_scale = getOption("notame.fill_scale_con"),
assay.type = NULL,
...
)Arguments
- object
a SummarizedExperiment or MetaboSet object
- all_features
logical, should all features be used? If FALSE (the default), flagged features are removed before visualization.
- center
logical, should the data be centered prior to PCA? (usually yes)
- scale
scaling used, as in
prep. Default is "uv" for unit variance- pca_method
the method used in PCA if there are missing values
- perplexity
the perplexity used in t-SNE
- fill
character, name of the column used for coloring the hexagons
- summary_fun
the function used to compute the value for each hexagon
- bins
the number of bins in x and y axes
- title, subtitle
the titles of the plot
- fill_scale
the fill scale as returned by a ggplot function
- assay.type
character, assay to be used in case of multiple assays
- ...
additional arguments passed to
Rtsne
Examples
data(example_set)
plot_tsne_hexbin(example_set, perplexity = 10)
#> Warning: Computation failed in `stat_summary_hex()`.
#> Caused by error in `compute_group()`:
#> ! The package "hexbin" is required for `stat_summary_hex()`.
