README.md

BSDE: Barycenter Single-cell RNAseq Differential Expression analysis for case-control studies

rstudio

Installation

# install.packages("devtools")
devtools::install_github("mqzhanglab/BSDE", build_vignettes = TRUE)

In case of missing dependency, the following R packages need to be installed from BioConductor:

if (!requireNamespace("BiocManager", quietly = TRUE))
    install.packages("BiocManager")
BiocManager::install("scater")
BiocManager::install("SingleCellExperiment")

Python dependency

Additionally, the package calls (via reticulate) Python module POT to compute optimal transport. To things work properly, POT should be pre-installed in a conda environment named r-reticulate. This can be set up automatically the first time BSDE is loaded.

But if that fails, you can install manually:

  1. First, make sure conda/miniconda is installed. You can follow the instructions here.

  2. Run the following in a terminal (line by line):

bash conda create --name r-reticulate python=3.7 conda activate r-reticulate conda install -c conda-forge pot

Still have problem loading?

If you are using RStudio, make sure the Python interpreter is not overridden to something else. In the PreferencesPython, select interpreter from the r-reticulate conda environment. Then, restart R and try library(BSDE). You should be able to see message "Python environment loaded".

rstudio

Documents



mqzhanglab/BSDE documentation built on March 20, 2022, 5:17 a.m.