This repository contains code scripts to reproduce analyses and figures in our paper "nnSVG for the scalable identification of spatially variable genes using nearest-neighbor Gaussian processes".
nnSVG is a method for computationally scalable identification of spatially variable genes (SVGs) in spatially-resolved transcriptomics data.
nnSVG is available as an R package from Bioconductor and GitHub.
More details including installation instructions and a tutorial are available on the Bioconductor and GitHub pages.
The code in this repository is organized into subdirectories in the analyses/ directory.
References to original datasets used in the analyses are provided in the manuscript.
Datasets can be downloaded as SpatialExperiment formatted data objects from the STexampleData Bioconductor package within the preprocessing scripts.
R objects (.RData files) containing source data to reproduce individual figures from the manuscript within the analysis and evaluation scripts are available from Figshare.
nnSVG is described in the following paper: