BayesSpace provides tools for clustering and enhancing the resolution of spatial gene expression experiments.
BayesSpace clusters a low-dimensional representation of the gene expression matrix, incorporating a spatial prior to encourage neighboring spots to cluster together. The method can enhance the resolution of the low-dimensional representation into "sub-spots", for which features such as gene expression or cell type composition can be imputed.
BayesSpace has been built and tested on the following operating systems:
- Linux: Ubuntu 18.04.4 LTS (Bionic Beaver)
- macOS: 10.14.6 (Mojave), 10.15.6 (Catalina)
- Windows: 10, Server 2012 R2 Standard
BayesSpace requires R 4.0+ and Bioconductor 3.12+. Specific package dependencies are defined in the package DESCRIPTION and are managed by the Bioconductor and devtools installers.
BayesSpace is available through Bioconductor.
# Install the Bioconductor package manager, if necessary
if (!requireNamespace("BiocManager", quietly = TRUE))
install.packages("BiocManager")
BiocManager::install("BayesSpace")
The development version can be installed via Bioconductor (see instructions on
using the devel
branch) or from
github with devtools.
# Install devtools, if necessary
if (!requireNamespace("devtools", quietly = TRUE))
install.packages("devtools")
devtools::install_github("edward130603/BayesSpace")
Installation, including compilation, should take no more than one minute.
Installing from source on macOS (such as when installing via
devtools::install_github()) requires
Fortran to compile the Rcpp
code.
Download links for the appropriate macOS versions can be found here:
- Mojave (10.14) and Catalina (10.15) - gfortran8.2
- Sierra (10.12) and High Sierra (10.13) - gfortran6.3
Additional details on installing the R compiler tools for Rcpp on macOS can be found in this blog post.
Note about homebrew: While gfortran is available via homebrew, we've encountered issues linking to its libraries after installation. We recommend installing directly from the GNU Fortran repo.
For an example of typical BayesSpace usage, please see our package vignette for a demonstration and overview of the functions included in BayesSpace.
Running the entire vignette takes approximately 5m30s on a Macbook Pro with a 2.0 GHz quad-core processor and 16 GB of RAM.