MBASEDMetaAnalysis: Generic function to perform standard meta analysis.

In MBASED: Package containing functions for ASE analysis using Meta-analysis Based Allele-Specific Expression Detection

Description

Generic function to perform standard meta analysis.

Usage

MBASEDMetaAnalysis(zValuesMat, zVariancesMat, alternative = "two.sided",
  checkArgs = FALSE)

Arguments

zValuesMat	matrix of z-values, on standard normal scale. Each row represents a specific genomic locus, while each column represents a set of observed values across loci (in practice, multiple columns represent different outcomes of simulations).
zVariancesMat	matrix of (estimated) variances of each z-value in zValuesMat. The interpretation of rows and columns is the same as for zValuesMat.
alternative	one of 'two.sided', 'greater', 'less'. DEFAULT: 'two.sided'.
checkArgs	single boolean specifying whether arguments should be checked for adherence to specifications. DEFAULT: FALSE

Details

MBASEDMetaAnalysis performs meta analysis calculations in a vectorized fashion. Input matrices zValuesMat and zVariancesMat have one column for each set of loci ('independent studies') to be combined, with each row corresponding to an individual locus. MBASEDMetaAnalysis uses meta analysis approach to combine values in each column of zValuesMat into a single column-specific value of z (using corresponding supplied variances to appropriately weight contributions of each individual z). The function reports the resulting averaged z values, together with corresponding standard deviations (standard errors), for fixed-effects setting (note: random effects are not meaningful in the context of SNVs in ASE). If the supplied matrices have a single row (only one locus), no meta-analysis is possible, and the original value and corresponding standard deviations are returned.

Value

a list with 5 elements:

hetPVal	a 1-row marix of heterogeneity p-values.
hetQ	a 1-row matrix of heterogeneity statistics.
fixedEffectsMeans	a 1-row matrix of column-specific fixed-effects meta analysis restults.
fixedEffectsSEs	a 1-row matrix of estimated SEs of fixed-effects meta analysis results.
pvalueFixed	a 1-row matrix of p-values for fixed-effects analysis.

Examples

set.seed(127000)
zVals1=rnorm(5, mean=rep(2,5), sd=sqrt(1:5))
zVals2=rnorm(5, mean=0, sd=1)+c(0,0,5,0,0) ## one outlier
MBASED:::MBASEDMetaAnalysis(zValuesMat=matrix(c(zVals1, zVals2), ncol=2), zVariancesMat=matrix(c(1:5, rep(1,5)), ncol=2), alternative='two.sided')

MBASED documentation built on Nov. 8, 2020, 5:53 p.m.