Plot spatial variable feature results

Visualize normalized results produced by RunSpatialVariableFeatures(). The summary view shows feature ranks and, when finite p- or q-values are stored, significance. Results without permutation statistics are shown without a significance size mapping or legend. The surface view reuses SpatialSpotPlot() to draw spatial expression for selected features. Unless explicitly overridden, surfaces inherit the saved assay, layer, image and coordinate columns, even if the object's default assay has changed.

Usage

SpatialVariableFeaturePlot(
  object = NULL,
  plot_type = c("summary", "surface", "combined"),
  features = NULL,
  nfeatures = 10,
  score_col = "score",
  assay = NULL,
  layer = NULL,
  image = NULL,
  overlay_image = TRUE,
  image.alpha = 1,
  coord.cols = c("col", "row"),
  flip.y = TRUE,
  pt.size = NULL,
  pt.alpha = 0.9,
  stroke = 0.1,
  palette = "Spectral",
  palcolor = NULL,
  legend.position = "right",
  theme_use = "theme_scop",
  theme_args = list(),
  combine = TRUE,
  nrow = NULL,
  ncol = NULL,
  byrow = TRUE,
  srt = NULL,
  image.scale = c("lowres", "hires")
)

Arguments

object

A Seurat object.

plot_type

Plot type: "summary", "surface", or "combined".

features

Features to plot. If NULL, top features from the stored spatial variable feature result are used.

nfeatures

Number of top features used when features = NULL.

score_col

Result column used for the summary x-axis.

assay, layer

Expression assay and layer for surfaces. When NULL, use the saved analysis settings; explicit values override them.

image

Image for surfaces. When NULL, use the saved analysis image.

overlay_image, image.alpha

Draw the spatial image beneath spots.

coord.cols

Metadata coordinate columns for surfaces. When omitted, use the saved analysis columns.

flip.y

Reverse the y axis for metadata or FOV coordinates without a raster image. The default FALSE agrees with spatial networks and cell boundaries. Raster-backed display coordinates already include their flip.

pt.size, pt.alpha

Point size and transparency. pt.size = NULL scales with sqrt(n) (minimum 0.3). Rasterized points keep at least a two-pixel radius at raster.dpi = c(512, 512) and scale with raster.dpi.

stroke

Point border width.

palette, palcolor

Palette name (thisplot::show_palettes) or custom colors.

legend.position

Legend position for surface plots.

theme_use

Theme name or function.

theme_args

Theme name or function, plus extra theme arguments.

combine, nrow, ncol, byrow

Combine plots with patchwork::patchwork. combine = FALSE returns a list of ggplots.

srt

Deprecated alias for object; supply exactly one of the two. It will be removed in scop 1.0.0.

image.scale

Image scale factor matching the raster stored in the selected image. Use "hires" for a hires raster; do not modify Seurat scale-factor slots.

Value

A ggplot or patchwork object.

Examples

data(visium_human_pancreas_sub)
spatial <- Seurat::NormalizeData(
  visium_human_pancreas_sub,
  assay = "Spatial",
  verbose = FALSE
)
spatial <- RunSpatialVariableFeatures(
  object = spatial,
  assay = "Spatial",
  nfeatures = 10,
  verbose = FALSE
)
SpatialVariableFeaturePlot(object = spatial, plot_type = "summary")