bilevelAnalysisClassic: Bi-level meta-analysis in conjunction with a classical...
In BLMA: BLMA: A package for bi-level meta-analysis

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

Perform a bi-level meta-analysis in conjunction with any of the classical hypothesis testing methods, such as t-test, Wilcoxon test, etc.

1 2	bilevelAnalysisClassic(x, y = NULL, splitSize = 5, metaMethod = addCLT, func = t.test, p.value = "p.value", ...)

`x`	a list of numeric vectors
`y`	an optional list of numeric vectors
`splitSize`	the minimum number of size in each split sample. splitSize should be at least 3. By default, splitSize=5
`metaMethod`	the method used to combine p-values. This should be one of addCLT (additive method [1]), fishersMethod (Fisher's method [5]), stoufferMethod (Stouffer's method [6]), max (maxP method [7]), or min (minP method [8])
`func`	the name of the hypothesis test. By default func=t.test
`p.value`	the component that returns the p-value after performing the test provided by the func parameter. For example, the function t-test returns the class "htest" where the component "p.value" is the p-value of the test. By default, p.value="p.value"
`...`	additional parameters for func

This function performs a bi-level meta-analysis for the lists of samples [1]. It performs intra-experiment analyses to compare the vectors in x agains the corresponding vectors in y using the function intraAnalysisClassic in conjunction with the test provided in func. For example, it compares the first vector in x with the first vector in y, the second vector in x with the second vector in y, etc. When y is null, then the comparisons are reduced to one-sample tests. After these comparisons, we have a list of p-values, one for each comparision. The function then combines these p-values to obtain a single p-value using metaMethod.

the combined p-value

Tin Nguyen and Sorin Draghici

[1] T. Nguyen, R. Tagett, M. Donato, C. Mitrea, and S. Draghici. A novel bi-level meta-analysis approach – applied to biological pathway analysis. Bioinformatics, 32(3):409-416, 2016.

intraAnalysisClassic, intraAnalysisGene, bilevelAnalysisGene

set.seed(1)
l1 <- lapply(as.list(seq(3)),FUN=function (x) rnorm(n=10, mean=1))
l1
# one-sample t-test
lapply(l1, FUN=function(x) t.test(x, alternative="greater")$p.value)
# combining the p-values of one-sample t-tests:
addCLT(unlist(lapply(l1, FUN=function(x) t.test(x, alter="g")$p.value)))
#Bi-level meta-analysis
bilevelAnalysisClassic(x=l1, alternative="greater")