SPARROWThe sparrow may be small, but it is fully equipped with all vital organs…
SPARROW is a method for subsampling large scale data to obtain a lightweight presentation such that most of the statistical power is preserved with respect to downstream analysis.
The name “sparrow” comes from the Chinese idiom, “the sparrow may be small, but it is fully equiped with all of the vital organs”. The letters in the acronym stand for: Submodular Power-Adaptive Reduction for Representative Downstream Workflows.
Installation
To install SPARROW from GitHub:
# Install from GitHub
install.packages("devtools")
devtools::install_github("kaishumason/SPARROW")SPARROW can be paired with spotGLM for fast and scalable spatial omic data analysis.
We also recommend for you to install spotGLM and check out the tutorials from there:
# Install from GitHub
devtools::install_github("kaishumason/spotGLM")Getting Started
Please follow these tutorials to get started on your data:
Demonstration on a MERFISH dataset
Demonstration on a simulated spatial-barcoding data set that requires deconvolution
Demonstration on a Visium HD data set
SPARROW was originally designed for large scale spatial transcriptomics data analysis, to be used in conjunction with the spotGLM package for identifying spatial signals. However, it can be used as a general purpose data selection tool.