stat.t2 {sma}R Documentation

Two-sample t-statistics

Description

Computes two-sample t-statistics for each gene in a multi-slide microarray experiment.

Usage

stat.t2(X, cl, x.ratio=FALSE, var.equal=TRUE, ...)

Arguments

X if x.ratio=F, X is a list containing two components. The first component is a matrix of log intensity ratios M=log_2 (R/G) and the second component is the average log intensities A = log_2(R*G)/2, such as the output from stat.ma. If x.ratio=T, X is a matrix of log expression ratios only. The rows of X correspond to genes and columns correspond to different hybridizations, that is different slides.
cl vector of class labels. Must consist of integers 1 and 2.
x.ratio logical flag: if TRUE, the matrix X contains only log intensity ratios, if FALSE, X is a list containing two components. The first component is a matrix of log expression ratios and the second component contains average log intensities A.
var.equal logical flag: if TRUE, the variances of the class 1 and class 2 parent populations are assumed equal.
... other parameters used in t.test.

Value

List containing the following components

t the two-sample t-statistic for each gene;
Num the numerator of the t-statistic for each gene, with names attribute "Num";
Denominator the denominator of the t-statistic for each gene, with names attribute "Den";
n1 number of class 1 observations used to calculate the t-statistic for each gene;
n2 number of class 2 observations used to calculate the t-statistics for each gene;
Average A if x.ratio=F, the average across all hybridizations of A = log_2(R*G)/2, with names attribute "A.bar", if x.ratio=T, NULL is returned.

Note

t2stat.func is called by stat.t2 and is not typically used on its own.

Author(s)

Sandrine Dudoit, sandrine@stat.berkeley.edu
Yee Hwa Yang, yeehwa@stat.berkeley.edu

References

D. Freedman, R. Pisani, and R. Purves. (1998). Statistics, 3rd ed. NewYork: W.W. Norton.

See Also

t2stat.func, plot.t2, plot.qq, t.test.

Examples

data(MouseArray)
## mouse.setup <- init.grid()
## mouse.data <- init.data() ## see \emph{init.data}
## mouse.lratio <- stat.ma(mouse.data, mouse.setup)
cl <- c(rep(1,3), rep(2,3))
mouse.t2 <- stat.t2(mouse.lratio, cl)

[Package sma version 0.5.15 Index]