ImputeNetParallel: Network-based parallel imputation

In anacarolinaleote/ADImpute: Adaptive Dropout Imputer (ADImpute)

Description

ImputeNetParallel implements network-based imputation in parallel

Usage

ImputeNetParallel(drop.mat, arranged, cores =
BiocParallel::bpworkers(BPPARAM), type = 'iteration', max.iter = 50,
BPPARAM = BiocParallel::SnowParam(type = "SOCK"))
#'

Arguments

drop.mat	matrix, logical; dropout entries in the data matrix (genes as rows and samples as columns)
arranged	list; output of ArrangeData
cores	integer; number of cores used for paralell computation
type	character; either 'iteration', for an iterative solution, or 'pseudoinv', to use Moore-Penrose pseudo-inversion as a solution.
max.iter	numeric; maximum number of iterations for network imputation. Set to -1 to remove limit (not recommended)
BPPARAM	parallel back-end to be used during parallel computation. See BiocParallelParam-class.

Value

matrix; imputation results incorporating network information

anacarolinaleote/ADImpute documentation built on May 18, 2021, 10:11 p.m.