ebCoexpressMeta {EBcoexpress}R Documentation

A function for DC meta-analysis that combines individual study analyses

Description

This function performs a DC meta-analysis, using the hyperparameter estimates obtained from individual-study DC analyses via one of the ebCoexpress series.

Usage

ebCoexpressMeta(DList, conditionsList, pattern, hpEstsList, controlOptions = list())

Arguments

DList

A list of the individual D matrices from the studies under consideration. They should have the same dimensions, corresponding to the same gene-pairs and the same conditions

conditionsList

A list of the individual conditions arrays from the studies under consideration

pattern

The appropriate output from ebPatterns()

hpEstsList

A list of the individual hyperparameter estimate objects from the studies under consideration, as outputted from the ebCoexpress series (or initializeHP(), the format is the same)

controlOptions

A list with many options for controlling execution:

applyTransform: Should Fisher's Z-transformation be applied? Defaults to TRUE

verbose: Controls auto-commenting; set to 0 to turn off comments

convtol: Convergence tolerance for the EM; default is 5e-04

enforceFloor: Should EC proportion never drop below 0.8? Default is TRUE

Details

Since the meta-analysis model assumes that each study has its own study-specific parameters, those parameters should be estimated using a single-study DC function (one of the other members of the ebCoexpress series); their outputs are used by the hpEstsList option. The EM is then run to determine the system-wide mixing proportions, which are used to compute meta posterior probabilities for all EC/DC classes

Value

The output is a list with two members, MODEL and POSTPROBS:

MODEL is a list containing an array MIX and a list HPS. MIX contains estimated mixing proportions for EC/DC classes. HPS is the inputed list of lists

POSTPROBS is a p-by-L matrix containing posterior probabilities of EC and DC over all L EC/DC classes. The EC posterior probabilities will always be in the first column (which should be fed into crit.fun() if using the soft threshold). Total posterior probabilities of DC for each gene pair are found by summing over the other L-1 columns (or taking 1 minus the first (EC) column)

Author(s)

John A. Dawson <jadawson@wisc.edu>

References

Dawson JA and Kendziorski C. An empirical Bayesian approach for identifying differential co-expression in high-throughput experiments. (2011) Biometrics. E-publication before print: http://onlinelibrary.wiley.com/doi/10.1111/j.1541-0420.2011.01688.x/abstract

Examples

data(fiftyGenes)
tinyCond <- c(rep(1,100),rep(2,25))
D <- makeMyD(fiftyGenes, tinyCond, useBWMC=TRUE)
set.seed(3)
initHP <- initializeHP(D, tinyCond)

D1 <- D
D2 <- D
DList <- list(D1, D2)
cond1 <- tinyCond
cond2 <- tinyCond
conditionsList <- list(cond1, cond2)
pattern <- ebPatterns(c("1,1","1,2"))
initHP1 <- initHP
initHP2 <- initHP
out1 <- ebCoexpressZeroStep(D1, cond1, pattern, initHP1)
out2 <- ebCoexpressZeroStep(D2, cond2, pattern, initHP2)
hpEstsList <- list(out1$MODEL$HPS, out2$MODEL$HPS)

metaResults <- ebCoexpressMeta(
    DList, conditionsList, pattern, hpEstsList)

[Package EBcoexpress version 1.38.0 Index]