voomgls_Symm: General Linear Model Using Voom Output corSymm correlction...

In ntyet/rmRNAseq: RNA-Seq Analysis for Repeated-Measures Data

General Linear Model Using Voom Output corSymm correlction structure

Description

This function run general linear model with corSymm correlation structure in function gls for all genes where the input data come from the output of voom.

Usage

voomgls_Symm(
  v,
  Subject,
  Time,
  ncores = 1,
  C.matrix = NULL,
  beta0 = NULL,
  print.progress = FALSE
)

Arguments

v	output of voom function.
Subject	a vector of subjects or experimental units.
Time	a vector of time points.
ncores	number of cores for embarrassingly parallel procedure. Default value of ncores is 1.
C.matrix	is a list of matrix Ci in testing H0: Ci*beta = 0.
beta0	vector of the hypothesized value of beta, usually, beta0 is a 0 vector. The default option beta0 = NULL means that beta0 is a vector of 0.
print.progress	logical indicator, T or F, to print the progress.

Value

a data frame has G rows (= number of genes) containing all outputs from glsSymm function, shrinkage estimates of error variances, and F-type test statistics calculated by teststat function.

Examples

data(res)
data(covset)
v <- res$ori.res$v
v$E <- v$E[1:20,]
v$weights <- v$weights[1:20,]
Subject <- covset$ear
Time <- covset$time
ncores <- 1
C.matrix <- list()
C.matrix[[1]] <- limma::makeContrasts(line2, levels = design)
C.matrix[[2]] <- limma::makeContrasts(time2, time6, time24, levels = design)
names(C.matrix) <- c("line2", "time")
beta0 <- NULL
print.progress <- FALSE
voomglsout <- rmRNAseq:::voomgls_Symm(v, Subject, Time, ncores, C.matrix, beta0, print.progress)

ntyet/rmRNAseq documentation built on July 7, 2023, 11:10 a.m.