SIEVEseq: Unified differential expression, variability, and skewness analyses using RNA-Seq data

Hongxiang Li and Tsung Fei Khang

SIEVEseq: A statistical method for building a unified framework for the simultaneous testing of differential expression, variability and skewness of genes using RNA-Seq data. This framework adopts a compositional data analysis approach to modelling RNA-Seq count data, applies the centered log-ratio (CLR) transformation to convert them into continuous variables, and uses a skew-normal distribution to model them.

Installation:

Install SAVER from CRAN by running:

install.packages("SIEVEseq")

library("SIEVEseq")

Install SIEVEseq from local source with

install.packages("SIEVEseq_0.0.0.tar.gz", repos=NULL, type="source")

Install SIEVE from GitHub with

if (!"devtools" %in% installed.packages()) { install.packages("devtools")}

devtools::install_github("Divo-Lee/SIEVEseq")

or

if (!"pak" %in% installed.packages()) { install.packages("pak")}

pak::pkg_install("Divo-Lee/SIEVEseq")

Dependencies:

SIEVEseq R package depends on the following packages: sn, stats, utils, ggplot2.