hfp: Function to construct a heatmap of the hidden variation in...

In svapls: Surrogate variable analysis using partial least squares in a gene expression study.

Description

The function hfp produces a plot of the PLS imputed estimate of the hidden variability in the data, derived from the optimal model, corresponding to an user-specified set of genes and subjects/samples.

Usage

hfp(obj, gen, ind, Y)

Arguments

obj	An svpls object.
gen	An user-specified set of genes.
ind	An user-specified set of subjects.
Y	A log transformed gene expression matrix with genes along the rows and subjects/samples along the columns.

Value

A heatmap of the hidden variability corresponding to the specified set of genes and subjects, attributable to the unknown subject-specific factors in the gene expression data.

Author(s)

Sutirtha Chakraborty, Somnath Datta and Susmita Datta.

References

Sutirtha Chakraborty, Somnath Datta and Susmita Datta. (2012) Surrogate Variable Analysis Using Partial Least Squares in Gene Expression Studies. Bioinformatics.

See Also

heatmap, fitModel, svpls

Examples

## Fitting the optimal ANCOVA model to the data gives:
data(hidden_fac.dat)
fit <- svpls(10,10,hidden_fac.dat,pmax = 5)

## Specifying the sets of genes and subjects
gen <- paste("g",c(1:15,50:65),sep="")
sub <- paste("S",c(1:5,11:17),sep="")

hfp(fit,gen,sub,hidden_fac.dat)

svapls documentation built on May 2, 2019, 9:34 a.m.