estimatecc: Estimate cell composition from DNAm data

In stephaniehicks/methylCC: Estimate the cell composition of whole blood in DNA methylation samples

Estimate cell composition from DNAm data

Description

Estimate cell composition from DNAm data

Usage

estimatecc(object, find_dmrs_object = NULL, verbose = TRUE,
  epsilon = 0.01, max_iter = 100, take_intersection = FALSE,
  include_cpgs = FALSE, include_dmrs = TRUE,
  init_param_method = "random", a0init = NULL, a1init = NULL,
  sig0init = NULL, sig1init = NULL, tauinit = NULL, demo = FALSE)

Arguments

object	an object can be a RGChannelSet, GenomicMethylSet or BSseq object
find_dmrs_object	If the user would like to supply different differentially methylated regions, they can use the output from the find_dmrs function to supply different regions to estimatecc.
verbose	TRUE/FALSE argument specifying if verbose messages should be returned or not. Default is TRUE.
epsilon	Threshold for EM algorithm to check for convergence. Default is 0.01.
max_iter	Maximum number of iterations for EM algorithm. Default is 100 iterations.
take_intersection	TRUE/FALSE asking if only the CpGs included in object should be used to find DMRs. Default is FALSE.
include_cpgs	TRUE/FALSE. Should individual CpGs be returned. Default is FALSE.
include_dmrs	TRUE/FALSE. Should differentially methylated regions be returned. Default is TRUE.
init_param_method	method to initialize parameter estimates. Choose between "random" (randomly sample) or "known_regions" (uses unmethyalted and methylated regions that were identified based on Reinus et al. (2012) cell sorted data.). Defaults to "random".
a0init	Default NULL. Initial mean methylation level in unmethylated regions
a1init	Default NULL. Initial mean methylation level in methylated regions
sig0init	Default NULL. Initial var methylation level in unmethylated regions
sig1init	Default NULL. Initial var methylation level in methylated regions
tauinit	Default NULL. Initial var for measurement error
demo	TRUE/FALSE. Should the function be used in demo mode to shorten examples in package. Defaults to FALSE.

Value

A object of the class estimatecc that contains information about the cell composition estimation (in the summary slot) and the cell composition estimates themselves (in the cell_counts slot).

Examples

# This is a reduced version of the FlowSorted.Blood.450k 
# dataset available by using BiocManager::install("FlowSorted.Blood.450k),
# but for purposes of the example, we use the smaller version 
# and we set \code{demo=TRUE}. For any case outside of this example for 
# the package, you should set \code{demo=FALSE} (the default). 

dir <- system.file("data", package="methylCC")
files <- file.path(dir, "FlowSorted.Blood.450k.sub.RData") 
if(file.exists(files)){
    load(file = files)

    set.seed(12345)
    est <- estimatecc(object = FlowSorted.Blood.450k.sub, demo = TRUE) 
    cell_counts(est)
 }   

stephaniehicks/methylCC documentation built on Nov. 6, 2022, 5:51 p.m.