The R package csrnaseq
performs a backward variable selection
procedure using pseudo-variables for RNA-seq differential expression
analysis (Nguyen and Nettleton 2024). The idea is to select the most
relevant covariates such that the false selection rate is below a
pre-specified threshold. The method is built upon the approach of Wu et
al. (2007). While Wu et al. (2007)’s method works for one response
variable, our method works for multiple response variables such as
RNA-seq data. The selected covariates are then included in differential
expression analysis using voom-limma
pipeline (Law et al. 2014). The
proposed method is implemented in function FSRAnalysisBS
.
csrnaseq
can be installed from
GitHub:
# install.packages("devtools")
devtools::install_github("ntyet/csrnaseq")
This is a basic example that shows how to use our method:
library(csrnaseq)
data(counts)
data(FixCov)
data(VarCov)
option <- "OWN"
B <- 2
m <- 2
alphamax <- 5
alpha0 <- 0.05
ncores <- 1
print.progress <- FALSE
saveall <- TRUE
FSRAnalysisBSOut <- FSRAnalysisBS(counts, FixCov, VarCov,
option, B, m, alphamax, alpha0,
ncores, print.progress, saveall)
names(FSRAnalysisBSOut)
-
Codes for generating six simulation scenarios are here and here. The outputs are here.
-
Codes for simulation are here and here. The outputs are here.
-
Codes for additional simulation to investigate covariates orthogonal to the primary variables that include:
-
Zebrafish RNA-seq dataset are available at here.
-
Codes for Zebrafish RNA-seq data analysis are here and here.
-
Codes for generating two simulation scenarios are here. The outputs are here.
-
Codes for simulation are here and here. The outputs are here.
Law, C. W., Chen, Y., Shi, W., and Smyth, G. K. (2014), “Voom: Precision weights unlock linear model analysis tools for RNA-seq read counts,” Genome Biology, 15, R29. https://doi.org/10.1186/gb-2014-15-2-r29.
Nguyen, Y., and Nettleton, D. (2024), “Identifying relevant covariates in RNA-seq analysis by pseudo-variable augmentation,” Journal of Agricultural, Biological and Environmental Statistics. https://doi.org/10.1007/s13253-024-00665-3.
Wu, Y., Boos, D. D., and Stefanski, L. A. (2007), “Controlling variable selection by the addition of pseudovariables,” Journal of the American Statistical Association, Taylor & Francis, 102, 235–243. https://doi.org/10.1198/016214506000000843.