Perform batch effect correction

Perform batch effect correction based on the function ComBat form R package sva. It requires that the "batches" in the data set are known. It uses either parametric or non-parametric empirical Bayes frameworks for adjusting data for batch effects.

Usage

batchEWAS(
  input,
  adjustVar = NULL,
  batch = NULL,
  plot = FALSE,
  par.prior = TRUE,
  mean.only = FALSE,
  ref.batch = NULL,
  parallel = FALSE,
  core = NULL
)

Arguments

• input:

  An R6 class integrated with all the information.

• adjustVar:

  (Optional) Names of the variate of interest and other covariates besides batch, with each name separated by a comma. Ensure that when correcting for batch effects, the effects of other factors are appropriately considered and adjusted for.Ensure there are no space. e.g. "cov1,cov2".

• batch:

  Name of the batch variable.

• plot:

  Logical. TRUE give prior plots with black as a kernel estimate of the empirical batch effect density and red as the parametric. Thr default is FALSE.

• par.prior:

  Logical. TRUE indicates parametric adjustments will be used, FALSE indicates non-parametric adjustments will be used.

• mean.only:

  Logical. Default to FALSE. If TRUE, ComBat only corrects the mean of the batch effect (no scale adjustment).

• ref.batch:

  (Optional) NULL If given, will use the selected batch as a reference for batch adjustment.

• parallel:

  Logical. Whether to enable parallel computing during batch effect correction. Default is FALSE.

• core:

  Integer. Number of CPU cores to use if parallel = TRUE. Default is NULL.

Value

input, An R6 class object integrating all information.

Examples

if (FALSE) { # \dontrun{
res <- initEWAS(outpath = "default")
res <- loadEWAS(input = res, ExpoData = "default", MethyData = "default")
res <- transEWAS(input = res, Vars = "cov1", TypeTo = "factor")
res <- batchEWAS(input = res, batch = "batch", par.prior=TRUE, ref.batch = NULL)
} # }