ThresholdBic: Methods for defining a bicluster

In rbentham/MCbiclust: Massive correlating biclusters for gene expression data and associated methods

Methods for defining a bicluster

Description

A bicluster is the fundamental result found using MCbiclust. These three functions are essential for the precise definition of these biclusters.

Usage

ThresholdBic(cor.vec, sort.order, pc1, samp.sig = 0)

PC1Align(gem, pc1, cor.vec, sort.order, bic)

ForkClassifier(pc1, samp.num)

Arguments

cor.vec	Correlation vector (output of CVEval()).
sort.order	Order of samples (output of SampleSort()).
pc1	PC1 values for samples (output of PC1VecFun).
samp.sig	Value between 0 and 1 determining number of samples in bicluster
gem	Gene expression matrix containing genes as rows and samples as columns.
bic	bicluster (output of ThresholdBic())
samp.num	Number of samples in the bicluster

Details

ThresholdBic() takes as its main inputs the correlation vector (output of CVEval()), sample ordering (output of SampleSort()), PC1 vector (output of PC1VecFun) and returns a list of the genes and samples which belong to the bicluster according to a certain level of significance.

PC1Align() is a function used once the bicluster has been found to ensure that the upper fork samples (those with higher PC1 values) correspond to those samples that have genes with positive correlation vector values up-regulated.

ForkClassifier() is a function used to classify which samples are in the upper or lower fork.

Value

Defined bicluster

Examples

data(CCLE_small)
data(Mitochondrial_genes)

mito.loc <- (row.names(CCLE_small) %in% Mitochondrial_genes)
CCLE.mito <- CCLE_small[mito.loc,]

set.seed(102)
CCLE.seed <- FindSeed(gem = CCLE.mito,
                      seed.size = 10,
                      iterations = 100,
                      messages = 1000)

CCLE.sort <- SampleSort(gem = CCLE.mito,seed = CCLE.seed,sort.length = 11)

# Full ordering are in Vignette_sort in sysdata.rda
CCLE.samp.sort <- MCbiclust:::Vignette_sort[[1]]

CCLE.pc1 <- PC1VecFun(top.gem = CCLE.mito,
                      seed.sort = CCLE.samp.sort,
                      n = 10)

CCLE.cor.vec <-  CVEval(gem.part = CCLE.mito,
                            gem.all = CCLE_small,
                            seed = CCLE.seed,
                            splits = 10)

CCLE.bic <- ThresholdBic(cor.vec = CCLE.cor.vec,sort.order = CCLE.samp.sort,
                         pc1 = as.numeric(CCLE.pc1))

CCLE.pc1 <- PC1Align(gem = CCLE_small, pc1 = CCLE.pc1,
                     cor.vec = CCLE.cor.vec ,
                     sort.order = CCLE.samp.sort,
                     bic =CCLE.bic)

CCLE.fork <- ForkClassifier(CCLE.pc1, samp.num = length(CCLE.bic[[2]]))

rbentham/MCbiclust documentation built on Feb. 5, 2024, 7:44 a.m.