emnormalCNV: Implement the EM algorithm

In CNVrd2: CNVrd2: a read depth-based method to detect and genotype complex common copy number variants from next generation sequencing data.

Description

This function is used to obtain the maximization likelihood estimation of normal mixture model by using the EM algorithm (Demster et al., 1977).

Usage

emnormalCNV(Object, ...)

Arguments

Object	An object of class clusteringCNVs.
...	Optional arguments

Value

loglk	Value of the likelihood function.
p	Proportions of groups.
m	Means of groups.
sigma	Standard deviations of groups.
count	A number of iteration to obtain convergence stage.
bic	See searchGroupCNVs.
z	Data frame of proportions of data in mixture components.

Note

In the package, the distance between two initial means of the two nearest neighbor groups was set groupDistance = 0.25 as a default value to obtain initial values (using the kmeans function in R).

Author(s)

Hoang Tan Nguyen, Tony R Merriman and MA Black. [e-mail]

References

Dempster, A. P., Laird, N. M., Rubin, D. B., 1977. Maximum likelihood from incomplete data via the em algorithm. Journal of the Royal Statistical Society. Series B (Methodological), 1-38.

See Also

searchGroupCNVs, groupCNVs

Examples

data(fcgr3bMXL)

sS <- resultSegment$segmentationScores
#########Histogram###########################
###View segmentation scores##################
hist(sS[, 1], 100)
############################################
##Number of components#######################
###Make an object of clusteringCNVs class######
objectCluster <- new("clusteringCNVs",
                     x = sS[, 1], k = 4, EV = TRUE)

set.seed(123)
copynumberGroups <- groupCNVs(Object = objectCluster)

CNVrd2 documentation built on Nov. 8, 2020, 5:30 p.m.