Skip to contents

Spatial Artifacts

The goal of SpatialArtifacts is to detect interior and edge artifacts, such as dry spots caused by incomplete reagent coverage or tissue handling, in spatial transcriptomics data. The package currently supports the 10x Genomics Visium and VisiumHD platforms.

If you experience any issues using the package or would like to make a suggestion, please open an issue on the GitHub repository.

To find more information, please visit the documentation website.

Key Features

  • Multi-platform support: Works on both standard 10x Visium and high-resolution Visium HD
  • Morphological detection: Uses raster-based focal transformations (fill, outline, star-pattern) to intelligently identify artifact clusters
  • Hierarchical classification: Categorizes artifacts into actionable groups (e.g., Large Edge Artifact, Small Interior Artifact)
  • Fast and efficient: Optimized with the terra R package for handling large datasets

Installation

You can install the latest version of SpatialArtifacts from Bioconductor with the following code:

if (!requireNamespace("BiocManager", quietly = TRUE)) {
    install.packages("BiocManager")
}

BiocManager::install("SpatialArtifacts")

You can install the development version of SpatialArtifacts from GitHub with:

# install.packages("pak")
pak::pak("CambridgeCat13/SpatialArtifacts")

Example

This is a basic example which shows you how to solve a common problem:

library(SpatialArtifacts)
library(SpatialExperiment)

# 1. Detect artifacts
# Option A: Standard Visium (Hexagonal Grid)
# Just specify platform = "visium"
spe <- detectEdgeArtifacts(spe, platform = "visium", qc_metric = "sum_umi")

# Option B: Visium HD (Square Grid)
# Specify platform = "visiumhd" AND the required resolution ("8um" or "16um")
# spe <- detectEdgeArtifacts(spe, platform = "visiumhd", resolution = "8um")

# 2. Classify results (Platform independent)
# Note: For Visium HD, remember to increase min_spots (e.g., min_spots = 400)
spe <- classifyEdgeArtifacts(spe)

# 3. View classification
table(spe$edge_artifact_classification)

Tutorials

A detailed tutorial is available in the package vignette from Bioconductor. A direct link to the tutorial / package vignette is available here.

Development tools

  • Continuous code testing is possible thanks to GitHub actions.
  • The documentation website is automatically updated thanks to BiocStyle::CRANpkg('pkgdown').
  • The documentation is formatted thanks to BiocStyle::CRANpkg('devtools') and BiocStyle::CRANpkg('roxygen2').
  • This package was developed using BiocStyle::Biocpkg('biocthis').

Contributors