Foldseq: Log fold change detection based on an empirical bayesian...

In cuiyingbeicheng/Foldseq: Log fold change detection based on an empirical bayesian method.

Description

Foldseq implements an empirical Bayesian method to substantially improve the power and accuracy in fold change detection.

Usage

Foldseq(ctrl, trt, cis.chr, cis.null.lfc = 0, trans.null.lfc = 0)

Arguments

ctrl	A data list for the control group including the gene expression matrix, gene id and the chromosome correspond to the gene.
trt	A data list for the treatment group including the gene expression matrix, gene id and the chromosome correspond to the gene.
cis.chr	the chromosome under treatment
cis.null.lfc	cut-off for testing cis genes that is on the treated chromosome
trans.null.lfc	cut-off for testing trans genes that is genes other that the cis genes

Details

Suppose the control group and treatment group have their population mean expression $mu^g_1$ and $mu^g_2$. The objective of our study is to detect if the ratio of $mu^g_2$ and $mu^g_1$ (i.e., fold change), is within or outside a region of interest, such as $mu^g_2/mu^g_1 >= d_0$, or $mu^g_2/mu^g_1 <= d_0$, or $d_1 <= mu^g_2/mu^g_1 <= d_2$, among others, with strong statistical evidence.

Value

cis.high	a table to summary the statistics of highly expressed genes located on the treated chromosome
cis.lower	a table to summary the statistics of lowly expressed genes located on the treated chromosome
cis.null	a table to summary the statistics of the non-differential genes located on the treated chromosome
trans.high	a table to summary the statistics of highly expressed genes located on the untreated chromosom
trans.lower	a table to summary the statistics of lowly expressed genes located on the untreated chromosome
trans.null	a table to summary the statistics of the non-differential genes located on the untreated chromosome

Examples

## An example to detect log fold change
data(smalldata)
Foldseq(ctrl=smalldata$Ctrl, trt=smalldata$Trt, cis.chr="chr1", cis.null.lfc = 0, trans.null.lfc = log2(1.5))

cuiyingbeicheng/Foldseq documentation built on May 18, 2020, 6:31 a.m.