Skip to contents

Render a cell plot as an interactive 3D widget

cell_plot_interactive.Rd

Builds a cell_plot() recipe and draws it as an interactive 3D scatter with one of two rendering backends, returned as an htmlwidget. Drag to rotate and scroll to zoom. Markers use the rendered colors resolved by build_cell_plot(), including illumination when requested, while legends are built from the unilluminated color scale metadata. Node sizes are converted from relative units to marker diameters in pixels.

Usage

cell_plot_interactive(object, renderer = c("rgl", "plotly"))

Arguments

object

A cell_plot recipe.

renderer

Rendering backend, "rgl" (default) or "plotly". The matching package must be installed.

Value

An htmlwidget. With renderer = "rgl" an rgl widget, with renderer = "plotly" a Plotly widget.

Details

Occlusion follows the scene camera, so markers closer to the current viewpoint appear in front. The arrange and depth mappings and cell_coord_rotate() are ignored. Panel grids support at most 10 rows and 20 columns.

renderer = "rgl" draws the scene with WebGL through rgl::rgl. The widget has no fixed size: it fills the IDE viewer, a Quarto or R Markdown page, or any other container and follows that container when it is resized. Panels in a grid share one camera and move together. Titles, facet strips, and legends are HTML drawn over the WebGL canvas, so they stay sharp at any resolution and their text can be selected. Their size follows the theme text size in points. Numeric color mappings get a colorbar and categorical mappings a discrete legend. Continuous sizes are grouped into at most 20 size bins. Hover labels and interactive legend filtering are not available.

renderer = "plotly" draws the scene with plotly::plotly. Each panel is a separate scene with its own camera, and hovering a marker shows its coordinates. Titles, facet strips, and legends are part of the Plotly layout.

See also

cell_plot(), print.cell_plot(), cell_plot_animate()

Examples

if (FALSE) { # interactive()
se <- ReadPNA_Seurat(minimal_pna_pxl_file())
se <- LoadCellGraphs(se, cells = colnames(se)[4], verbose = FALSE) |>
  ComputeLayout(layout_method = "spectral")

cell_graph <- CellGraphs(se)[[4]]

layout_data <- FetchLayoutData(cell_graph, vars = "CD82", layout_method = "spectral_3d")

# WebGL scene with rgl
cell_plot(layout_data, color = CD82) |>
  cell_plot_interactive()

# The same recipe drawn with Plotly
cell_plot(layout_data, color = CD82) |>
  cell_plot_interactive(renderer = "plotly")
}