M3D (version 1.6.2)

M3D_Para: Computes the components of the M3D test-statistic over all regions for all sample-pairs.

Description

Parallel implementation of M3D_Wrapper function. Returns the two components of the M3D test-statistic - the MMD (Gretton et al. 2006) for the full data and the coverge only data, respectively - for all regions and all samples pairs, as a matrix.

Usage

M3D_Para(rrbs, CpGs, overlaps, num.cores = NaN)

Arguments

rrbs

An rrbs object containing methylation and coverage data as created using the BiSeq pacakge

CpGs

A GRanges object detailing the testing regions.

overlaps

The overlaps between the list of testing regions and the methylation data. This is obtained using the function findOverlaps(CpGs,rrbs) for a GRanges object CpGs detailing the testing regions.

num.cores

Integer giving the number or cores to use. Defaults to the maximum available

Value

This returns the two components of the M3D test-statistic for each region over all sample pairs as a matrix. Subtracting them gives the M3D test-statistic. This is processed with the function pvals.

References

Gretton, A., Borgwardt, K. M., Rasch, M., Scholkopf, B., Smola, A. J. (2006). A kernel method for the two-sample-problem. In Advances in neural information processing systems (pp. 513-520).

Examples

data(rrbsDemo)
data(CpGsDemo)
M3D_list <- M3D_Para(rrbsDemo,CpGsDemo)
head(M3d_list$Full-M3D_list$Coverage)