fastROGUE enables accelerated computation on complex data by utilizing parallel processing. The main update allows the rogue function to specify ncores for parallel computation.
rogue(
expr,
labels,
samples,
platform = NULL,
k = NULL,
min.cell.n = 10,
remove.outlier.n = 2,
span = 0.9,
r = 1,
filter = F,
min.cells = 10,
min.genes = 10,
mt.method = "fdr",
ncores = 4, # <--- This one!
fix_to_numbers = T
)Installing fastROGUE
To install fastROGUE, run:
if (!requireNamespace("devtools", quietly = TRUE)) install.packages("devtools")
devtools::install_github("ZhimingYe/fastROGUE")
Origin readme:
Often, it is not even clear whether a given cluster is uniform in unsupervised scRNA-seq data analyses. Here, we proposed the concept of cluster purity and introduced a conceptually novel statistic, named ROGUE, to examine whether a given cluster is a pure cell population.
Installing dependency package
Before installing ROGUE, the “tidyverse” package should be installed first:
install.packages("tidyverse")
Installing ROGUE
To install ROGUE, run:
if (!requireNamespace("devtools", quietly = TRUE)) install.packages("devtools")
devtools::install_github("PaulingLiu/ROGUE")
For more details and basic usage see following tutorials: 1. Guided Tutorial (It takes a few seconds to load the HTML file)
The scripts for producing all the quantitative results in our manuscript can be found in scripts.
If you use ROGUE in your research, please considering citing: - Liu et al., Nature Communications 2020
Please contact us:
Baolin Liu: [email protected]{.email}
Zemin Zhang: [email protected]{.email}
©2019 Baolin Liu, Chenwei Li. Zhang Lab. All rights reserved.