CoMethAllRegions: Extract contiguous co-methylated genomic regions from a list...

In lissettegomez/coMethDMR: Accurate identification of co-methylated and differentially methylated regions in epigenome-wide association studies

Description

Extract contiguous co-methylated genomic regions from a list of pre-defined genomic regions

Usage

CoMethAllRegions(
  dnam,
  betaToM = FALSE,
  method = c("pearson", "spearman"),
  rDropThresh_num = 0.4,
  minCpGs = 3,
  arrayType = c("450k", "EPIC"),
  CpGs_ls,
  file = NULL,
  returnAllCpGs = FALSE,
  output = c("CpGs", "dataframe"),
  nCores_int = 1L,
  ...
)

Arguments

dnam	matrix (or data frame) of beta values, with row names = CpG IDs, column names = sample IDs. This is typically genome-wide methylation beta values.
betaToM	indicates if converting methylation beta values to mvalues
method	method for computing correlation, can be "spearman" or "pearson"
rDropThresh_num	threshold for min correlation between a cpg with sum of the rest of the CpGs
minCpGs	minimum number of CpGs to be considered a "region". Only regions with more than minCpGs will be returned.
arrayType	Type of array, can be "450k" or "EPIC"
CpGs_ls	list where each item is a character vector of CpGs IDs. This should be CpG probes located closely on the array.
file	an RDS file with clusters of CpG locations (i.e. CpGs located closely to each other on the genome). This file can be generated by the WriteCloseByAllRegions function.
returnAllCpGs	When there is not a contiguous comethylated region in the inputting pre-defined region, returnAllCpGs = 1 indicates outputting all the CpGs in the input regions, while returnAllCpGs = 0 indicates not returning any CpG.
output	a character vector of CpGs or a dataframe of CpGs along with rDrop info
nCores_int	Number of computing cores to be used when executing code in parallel. Defaults to 1 (serial computing).
...	Dots for additional arguments passed to the cluster constructor. See CreateParallelWorkers for more information.

Details

There are two ways to input genomic regions for this function: (1) use CpGs_ls argument (2) use file argument

examples of these files are in /inst/extdata/ folder of the package.

Value

When output = "dataframe" is selected, returns a list of data frames, each with CpG (CpG name), Chr (chromosome number), MAPINFO (genomic position), r_drop (correlation between the CpG with rest of the CpGs), keep (indicator for co-methylated CpG), keep_contiguous (index for contiguous comethylated subregions).

When output = "CpGs" is selected, returns a list, each item is a list of CpGs in the contiguous co-methylated subregion.

Examples

   data(betasChr22_df)


   CpGisland_ls <- readRDS(
     system.file(
       "extdata",
       "CpGislandsChr22_ex.RDS",
       package = 'coMethDMR',
       mustWork = TRUE
     )
   )

   coMeth_ls <- CoMethAllRegions (
     dnam = betasChr22_df,
     betaToM = TRUE,
     method = "pearson",
     CpGs_ls = CpGisland_ls,
     arrayType = "450k",
     returnAllCpGs = FALSE,
     output = "CpGs"
   )

lissettegomez/coMethDMR documentation built on April 25, 2021, 1:10 p.m.