PointScoreCalc: Calculate PointScore

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

Description

Using two gene sets that are represented of a known bicluster (one gene set being up regulated while other gene set is down regulated), samples are scored based on how well they match the known regulation of the bicluster.

Usage

PointScoreCalc(gene.expr.matrix, gene.loc1, gene.loc2)

Arguments

gene.expr.matrix	Gene expression matrix with genes as rows and samples as columns
gene.loc1	Location of the rows containing the genes in gene set 1 within the gene expression matrix
gene.loc2	Location of the rows containing the genes in gene set 2 within the gene expression matrix

Details

The PointScore of a sample can be directly compared to the PC1 value. The PointScore is typically used to identify samples related to the upper/lower fork of a bicluster without running the complete main MCbiclust pipeline on a dataset.

Value

Vector of point scores for each sample in the gene expression matrix

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.hicor.genes <- as.numeric(HclustGenesHiCor(CCLE.mito,
                                                CCLE.seed,
                                                cuts = 8))

CCLE.cor.mat <- cor(t(CCLE.mito[CCLE.hicor.genes,CCLE.seed]))

gene.set1 <- labels(as.dendrogram(hclust(dist(CCLE.cor.mat)))[[1]])
gene.set2 <- labels(as.dendrogram(hclust(dist(CCLE.cor.mat)))[[2]])

gene.set1.loc <- which(row.names(CCLE.mito) %in% gene.set1)
gene.set2.loc <- which(row.names(CCLE.mito) %in% gene.set2)

ps.vec <- PointScoreCalc(CCLE.mito,gene.set1.loc,gene.set2.loc)

cor(ps.vec[CCLE.samp.sort], CCLE.pc1)
plot(ps.vec[CCLE.samp.sort])
plot(CCLE.pc1)

compmedlab/MCbiclust documentation built on March 9, 2020, 12:14 a.m.