Step: UMAP

Perform UMAP analysis using glystats::gly_umap() and plot a UMAP plot using glyvis::plot_umap(). Note that the result of UMAP largely depends on the n_neighbors parameter. Usually it's a trial-and-error process to find the best value iteratively. If you are not satisfied with the result, manually call glyvis::plot_umap() with different n_neighbors values to find the best one.

Usage

step_umap(
  on = "exp",
  n_neighbors = 15,
  n_components = 2,
  plot_width = 5,
  plot_height = 5,
  ...
)

Arguments

on: Name of the experiment to run UMAP on. Can be "exp", "sig_exp", "trait_exp", "sig_trait_exp", "motif_exp", "sig_motif_exp".
n_neighbors: Number of neighbors to consider for each point. Default is 15.
n_components: Number of output dimensions. Default is 2.
plot_width: Width of the plot in inches. Default is 5.
plot_height: Height of the plot in inches. Default is 5.
...: Additional arguments passed to uwot::umap().

Value

A glysmith_step object.

Details

Data required:

exp (if on = "exp"): The experiment to perform UMAP on
trait_exp (if on = "trait_exp"): The trait experiment to perform UMAP on
motif_exp (if on = "motif_exp"): The motif experiment to perform UMAP on

Data generated (with suffixes):

umap: The UMAP result

Plots generated (with suffixes):

umap: The UMAP plot

AI Prompt

This section is for AI in inquire_blueprint() only.

Include this step only when the user explicitly asks for UMAP.

Examples

step_umap()
#> <step "step_umap()"> UMAP
step_umap(n_neighbors = 15)
#> <step "step_umap(n_neighbors = 15)"> UMAP