single.chip.enrichment: Calculate enrichment of a list of genesets in an array
In hidelab/pathprint: Pathway fingerprinting for analysis of gene expression arrays

Description Usage Arguments Details Value Author(s) References See Also Examples

Function to assess enrichment of gene sets in an array or matrix of arrays using various summary statistics

single.chip.enrichment(exprs,
    geneset,
    transformation = "rank",
    statistic = "mean",
    normalizedScore = FALSE,
    progressBar = TRUE)

`exprs`	An expression matrix, rownames correspond to gene ids used in the list of genesets
`geneset`	list of pathways or genesets over which to assess statistic
`transformation`	Initial transformation applied to each column of exprs, can be one of "rank", "squared.rank" or "log.rank"
`statistic`	Summary statistic to be applied, either "mean" or "median"
`normalizedScore`	Logical. If statistic = "mean" and normalizedScore = TRUE, option to calculate a parametric significance score based on the expected distribution of scores. Other summary statistics currently not supported
`progressBar`	Logical. Shows progress of script, good to check running okay, set to FALSE for possible faster running

This is the worker function for exprs2fingerprint, in conjuction with an exprs based on Entrez Gene IDs and the standard pathprint genesets e.g. pathprint.Hs.gs. The (un-normalized) results are passed onto thresholdFingerprint to produce the Pathway Fingerprint scores

Matrix containing pathway enrichment scores for each sample in the exprs input matrix. Rownames are genesets and colnames are the columns of the exprs matrix.

Gabriel Altschuler

Altschuler, G. M., O. Hofmann, I. Kalatskaya, R. Payne, S. J. Ho Sui, U. Saxena, A. V. Krivtsov, S. A. Armstrong, T. Cai, L. Stein and W. A. Hide (2013). "Pathprinting: An integrative approach to understand the functional basis of disease." Genome Med 5(7): 68.

exprs2fingerprint

require(pathprintGEOData)
library(SummarizedExperiment)

# load  the data
data(SummarizedExperimentGEO)

# Compare continuous pathway enrichment values to Pathway Fingerprint scores

# Use ALL dataset as an example

require(ALL)
data(ALL)
annotation(ALL)

ds = c("chipframe","pathprint.Hs.gs","genesets","platform.thresholds")
data(list = ds)


# The chip used was the Affymetrix Human Genome U95 Version 2 Array
# The correspending GEO ID is GPL8300

# Analyze patients with ALL1/AF4 and BCR/ABL translocations
ALL.eset <- ALL[, ALL$mol.biol %in% c("BCR/ABL", "ALL1/AF4")]
ALL.exprs<-exprs(ALL.eset)

patient.type<-as.character(ALL$mol.biol[
    ALL$mol.biol %in% c("BCR/ABL", "ALL1/AF4")])

# Process fingerprints
ALL.fingerprint<-exprs2fingerprint(exprs = ALL.exprs,
    platform = "GPL8300",
    species = "human",
    progressBar = TRUE
    )

color.map <- function(mol.biol) {
    if (mol.biol=="ALL1/AF4") "#00FF00" else "#FF00FF"
    }
patientcolors <- sapply(ALL$mol.biol[
    ALL$mol.biol %in% c("BCR/ABL", "ALL1/AF4")],
    function(x){
    if (x == "ALL1/AF4") "#00FF00" else "#FF00FF"
    })


# define list of differentially activated pathways between the two groups
signif.pathways<-diffPathways(ALL.fingerprint,
    fac = patient.type,
    threshold = 0.6)

# draw heatmap
heatmap(ALL.fingerprint[signif.pathways,],
    ColSideColors = patientcolors,
    col = c("blue", "white", "red"),
    scale = "none", mar = c(10,20),
    cexRow = 0.75)
title(sub = "Pathways differentially activated in patients
    with ALL1/AF4 (green) and BCR/ABL(purple) translocations",
        cex.sub = 0.75)

######
# Compare to continous values
ALL.exprs.entrez <- customCDFAnn(ALL.exprs, chipframe$GPL8300$ann)
ALL.enrichment <- single.chip.enrichment(exprs = ALL.exprs.entrez,
    geneset = pathprint.Hs.gs,
    transformation = "squared.rank",
    statistic = "mean",
    normalizedScore = FALSE,
    progressBar = TRUE
    )

heatmap(ALL.enrichment[signif.pathways,],
    ColSideColors = patientcolors,
    col = colorRampPalette(c("blue", "white", "red"))(100),
    scale = "row", mar = c(10,20),
    cexRow = 0.75)
title(sub = "Continuous pathway enrichment scores for patients
    with ALL1/AF4 (green) and BCR/ABL(purple) translocations",
        cex.sub = 0.75)