/UCRSI

A clustering method Unifying cell-type recognition and subtypes identification for tumor heterogeneity analysis

Primary LanguageC++GNU General Public License v3.0GPL-3.0

UCRSI

A clustering method Unifying cell-type recognition and subtypes identification for tumor heterogeneity analysis


1. Install R dependencies package

install.packages("devtools")
install.packages("Rcpp")
install.packages("RcppArmadillo")
install.packages("Seurat")
install.packages("ggplot2")
install.packages("Matrix")
install.packages("reticulate")
install.packages("umap")
devtools::install_github("tnagler/RcppThread")

2. Python dependency

UCRSI relies on two Python packages, scikit-learn and pyamg, to complete the clustering process. In R, we use the reticulate package to call it. It is recommended to create a separate virtual environment using virtualenv. After configuring the environment, you need to change the python_env variable in the main.R file to the path of the environment.

Example

See main.R