ImputeNetwork: Network-based imputation

In ADImpute: Adaptive Dropout Imputer (ADImpute)

Description

Network-based imputation

Usage

ImputeNetwork(data, net.coef = NULL,
cores = BiocParallel::bpworkers(BPPARAM),
BPPARAM = BiocParallel::SnowParam(type = "SOCK"),
type = 'iteration', write = FALSE, ...)

Arguments

data	matrix with entries equal to zero to be imputed, normalized and log2-transformed (genes as rows and samples as columns)
net.coef	matrix; network coefficients.
cores	integer; number of cores to use
BPPARAM	parallel back-end to be used during parallel computation. See BiocParallelParam-class.
type	character; either 'iteration', for an iterative solution, or 'pseudoinv', to use Moore-Penrose pseudo-inversion as a solution.
write	logical; should a file with the imputation results be written?
...	additional arguments to ImputeNetParallel

Details

Imputes dropouts using a gene regulatory network trained on external data, as provided in net.coef. Dropout expression values are estimated from the expression of their predictor genes and the network coefficients.

Value

matrix; imputation results incorporating network information

See Also

ImputeNetParallel

ADImpute documentation built on Nov. 8, 2020, 5:30 p.m.