cmarrt.ma: Compute moving average statistics by incorporating the...

In Starr: Simple tiling array analysis of Affymetrix ChIP-chip data

Description

This function extends the moving average approach by incorporating the correlation structure. It also outputs the p-values of the standardized moving average statistics under the Gaussian approximation.

Usage

cmarrt.ma(eSet, probeAnno, chr=NULL, M=NULL,frag.length,window.opt='fixed.probe')

Arguments

eSet	ExpressionSet containing the normalized ratio
probeAnno	probeAnno object with mapping
chr	which chromosome should be analysed? If chr==NULL, all chromosome in the probeAnno object are taken.
M	rough estimate of the percentage of bound probes. If unknown, leave it NULL.
frag.length	average fragment length from sonication.
window.opt	option for sliding window, either "fixed.probe" or "fixed.gen.dist". Default is 'fixed.probe'.

Details

Computation using window.opt = "fixed.probe" calculates the moving average statistics within a fixed number of probes and is more efficient. Use this option if the tiling array is regular with approximately constant resolution. window.opt="fixed.gen.dist" computes the moving average statistics over a fixed genomic distance.

Value

data.sort	datafile sorted by genomic position.
ma	unstandardized moving average(MA) statistics.
z.cmarrt	standardized MA under correlation structure.
z.indep	standardized MA under independence (ignoring correlation structure).
pv.cmarrt	p-values of probes under correlation.
pv.indep	p-values of probes under independence (ignoring correlation structure).

Note

The p-values are obtained under the Gaussian approximation. Therefore, it is important to check the normal quantile-quantile plot if the Gaussian approximation is valid. The function also outputs the computation under independence (ignoring the correlation structure) for comparisons.

Author(s)

Pei Fen Kuan, Adam Hinz

References

P.F. Kuan, H. Chun, S. Keles (2008). CMARRT: A tool for the analysiz of ChIP-chip data from tiling arrays by incorporating the correlation structure. Pacific Symposium of Biocomputing13:515-526.

See Also

plotcmarrt,cmarrt.peak

Examples

# dataPath <- system.file("extdata", package="Starr")
# bpmapChr1 <- readBpmap(file.path(dataPath, "Scerevisiae_tlg_chr1.bpmap"))

# cels <- c(file.path(dataPath,"Rpb3_IP_chr1.cel"), file.path(dataPath,"wt_IP_chr1.cel"), 
# 	file.path(dataPath,"Rpb3_IP2_chr1.cel"))
# names <- c("rpb3_1", "wt_1","rpb3_2")
# type <- c("IP", "CONTROL", "IP")
# rpb3Chr1 <- readCelFile(bpmapChr1, cels, names, type, featureData=TRUE, log.it=TRUE)

# ips <- rpb3Chr1$type == "IP"
# controls <- rpb3Chr1$type == "CONTROL"

# rpb3_rankpercentile <- normalize.Probes(rpb3Chr1, method="rankpercentile")
# description <- c("Rpb3vsWT")
# rpb3_rankpercentile_ratio <- getRatio(rpb3_rankpercentile, ips, controls, description, fkt=median, featureData=FALSE)

# probeAnnoChr1 <- bpmapToProbeAnno(bpmapChr1)
# peaks <- cmarrt.ma(rpb3_rankpercentile_ratio, probeAnnoChr1, chr=NULL, M=NULL,250,window.opt='fixed.probe')

Starr documentation built on April 28, 2020, 7:52 p.m.