Skip to contents

Launch an Interactive ROI Annotation App for Seurat Spatial Data

Source: R/ROISelector.R
SelectROIs.Rd

This function opens a Shiny app to allow users to manually annotate Regions of Interest (ROIs) on spatial transcriptomics or metabolomics data stored in a Seurat object. Users can interactively select regions using lasso selection, assign custom names to each ROI, and save the results as binary (0/1) metadata columns in the Seurat object.

Usage

SelectROIs(seurat_obj, image = "fov")

Arguments

seurat_obj

A Seurat object containing spatial coordinates and metadata.

image

A character string specifying the spatial image source used by GetTissueCoordinates(). Default is "fov". Set to "VisiumV2" for Visium-style spatial data.

Value

A modified Seurat object with added metadata columns for each saved ROI (1 = inside ROI, 0 = outside).

Details

The app includes options to:

  • Choose a metadata column to display.

  • Adjust the spot size for display.

  • Use lasso to select spots and name each ROI.

  • Save each ROI to the Seurat object metadata.

  • Export the final Seurat object upon clicking "Finish".

Spatial selection uses sf geometry tools. The plot is rendered using plotly.