plotObsQC: plotObsQC

In lmweber/ggspavis: Visualization functions for spatial transcriptomics data

plotObsQC

Description

Plotting functions for spatial transcriptomics data.

Usage

plotObsQC(
  spe,
  plot_type = c("histogram", "scatter", "spot", "violin"),
  x_coord = NULL,
  y_coord = NULL,
  x_metric = NULL,
  y_metric = NULL,
  x_threshold = NULL,
  y_threshold = NULL,
  trend = TRUE,
  marginal = TRUE,
  annotate = NULL,
  in_tissue = NULL,
  legend_point_size = 3,
  n_bins = 100,
  point_size = 0.3,
  y_reverse = TRUE
)

plotQC(...)

plotSpotQC(...)

Arguments

spe	Input data, assumed to be a SpatialExperiment or SingleCellExperiment object.
plot_type	Type of QC plot. Options are "histogram", "scatter", "spot", and "violin". See Details for additional details.
x_coord	Name of column in spatialCoords (for a SpatialExperiment input object) or colData (for a SingleCellExperiment input object) containing x coordinates. Default = NULL (for a SpatialExperiment, the first column of spatialCoords will be selected in this case). Used for spot plots.
y_coord	Name of column in spatialCoords (for a SpatialExperiment input object) or colData (for a SingleCellExperiment input object) containing y coordinates. Default = NULL (for a SpatialExperiment, the second column of spatialCoords will be selected in this case). Used for spot plots.
x_metric	Name of column in colData containing QC metric to plot on x-axis. Required for histograms, scatter plots, and violin plots.
y_metric	Name of column in colData containing QC metric to plot on y-axis. Required for histograms, scatter plots, and violin plots.
x_threshold	QC filtering threshold on x-axis metric to highlight with vertical line. Default = NULL. Optional argument used for scatter plots.
y_threshold	QC filtering threshold on y-axis metric to highlight with horizontal line. Default = NULL. Optional argument used for scatter plots.
trend	Whether to show smoothed trend line (loess). Default = TRUE. Optional argument used for scatter plots.
marginal	Whether to show marginal histograms. Default = TRUE. Optional argument used for scatter plots.
annotate	Name of column in colData identifying selected spots that do not meet QC filtering thresholds, which will be highlighted on a histogram, spot plot, or violin plot. Default = NULL. Optional argument used for histograms, spot plots, and violin plots.
in_tissue	Name of column in colData identifying spots over tissue (e.g. "in_tissue" for 10x Genomics Visium datasets). If this argument is provided, only spots over tissue will be shown. Default = NULL. Optional argument used for spot plots.
legend_point_size	Legend point size. Default = 3. Optional argument used for spot plots.
n_bins	Number of bins for histograms. Default = 100. Optional argument used for histograms.
point_size	Point size. Default = 0.3. Optional argument for scatter plots, spot plots, and violin plots. Suggested values: 0.5 for scatter plots, 0.3 for spot plots, 0.1 for violin plots.
y_reverse	Whether to reverse y coordinates. This is usually required for 10x Genomics Visium datasets when using the default coordinate values. Default = TRUE. Set to FALSE if not needed, e.g. for other platforms. Optional argument used for spot plots.
...	Not used.

Details

Function to create observation-level quality control (QC) plots for spatial transcriptomics data.

The following types of observation-level (i.e. spot-level or cell-level) QC plots are available:

• Histogram (plot_type = "histogram") for a single QC metric, e.g. number of UMI counts per spot. For number of counts per spot, the histogram can optionally highlight selected spots, e.g. spots with low library size.

• Scatter plot (plot_type = "scatter") comparing two QC metrics, e.g. number of detected features vs. number of cells per spot, with optional horizontal and vertical lines highlighting QC filtering thresholds.

• Spot plot (plot_type = "spot") showing spots in spatial x-y coordinates, e.g. highlighting selected spots that do not meet filtering thresholds.

• Violin plot (plot_type = "violin") for a single QC metric, e.g. number of UMI counts per spot. For number of counts per spot, the violin plot can optionally highlight selected spots, e.g. spots with low library size.

For feature-level (i.e. gene-level) QC plots, see plotFeatureQC.

Value

Returns a ggplot object, which may be further modified using ggplot functions.

Author(s)

Lukas M. Weber and Yixing E. Dong

Examples

library(STexampleData)
spe <- Visium_humanDLPFC()

colData(spe)$sum <- colSums(counts(spe))
colData(spe)$low_libsize <- colData(spe)$sum < 400

plotObsQC(spe, plot_type = "histogram", x_metric = "sum", annotate = "low_libsize")
plotObsQC(spe, plot_type = "scatter", x_metric = "sum", y_metric = "cell_count")
plotObsQC(spe, plot_type = "spot", annotate = "low_libsize", in_tissue = "in_tissue")
plotObsQC(spe, plot_type = "violin", x_metric = "sum", annotate = "low_libsize")

lmweber/ggspavis documentation built on July 17, 2025, 3:18 a.m.