lmc.annotation.enrichment: lmc.annotation.enrichment

In lutsik/MeDeCom: Decomposition of heterogeneous methylomes

lmc.annotation.enrichment

Description

This function performs enrichment analysis for various genomic locations including chromosomes, and different function categories defined by the Ensembl regulatory build

Usage

lmc.annotation.enrichment(
  medecom.result,
  annotation.filter = NULL,
  anno.data,
  K = NULL,
  lambda = NULL,
  cg_subset = NULL,
  diff.threshold = 0.5,
  reference.computation = "median",
  comp.lmcs = NULL,
  type = "hypo",
  assembly = "hg19"
)

Arguments

medecom.result	An object of type MeDeComSet-class or the location of an .RData file, where such an object is stored.
annotation.filter	A numeric vector specifying the sites that have been removed from rnb.set in a preprocessing step (e.g. coverage filtering) or a path to an .RData file.
anno.data	The original RnBSet-class object containing methylation, sample meta and annotation information or a path to a directory stored by save.rnb.set or a data.frame containing CpG annotations (ann_C)
K	The number of LMCs specified for the MeDeCom run.
lambda	The lambda parameter selected.
cg_subset	The index of the selection strategy employed (e.g. most variable CpGs).
diff.threshold	The difference cutoff between median methylation in the remaining LMCs and the LMC of interest used to call a CpG differentially methylated. The higher this value, the more conservative the selection.
reference.computation	Metric used to set the reference on the remaining LMCs to determine hyper- and hypomethylated sites. Can be either "median" (default), "mean", or "lmcs" (comp.lmcs argument needs to be provided).
comp.lmcs	Numeric vector containing two numbers representing the LMCs that should be compared to one another.
type	Which direction is to be tested for enrichment. Can be one of "hypo", "hyper", or "differential"
assembly	The assembly used. Needs to be one of "hg19", "hg38" or "mm10". Does not need to be specified, if rnb.set is a RnBSet-class

Value

A data frame with four columns:

LMC

The LMC analyzed

annotation

The annotation used. Can either be chrXY for enrichments on different chromosmes or different functional categories in the Ensembl regulatory build

p.value

The p-value computing using Fisher's exact test for enrichment

OR

The odds ratio for enrichment

Author(s)

Michael Scherer

lutsik/MeDeCom documentation built on July 15, 2024, 4:51 a.m.