getPcs: Dimensionality reduction by PCA

In JEFworks/MUDAN: Multi-sample Unified Discriminant ANalysis

Description

Dimensionality reduction using PCA by computing principal components using highly variable genes

Usage

getPcs(mat, nGenes = min(nrow(mat), 1000), nPcs = 100, verbose = TRUE,
  ...)

Arguments

mat	Variance normalized gene expression matrix.
nGenes	Number of most variable genes. (default: 1000)
nPcs	Number of principal components. (default: 100)
verbose	Verbosity (default: TRUE)
...	Additional parameters to pass to irlba

Value

Matrix with columns as cells and rows as principal component eigenvectors.

Examples

{
data(pbmcA)
cd <- pbmcA[, 1:500]
mat <- cleanCounts(cd)
mat <- normalizeVariance(mat)
pcs <- getPcs(mat)
}

JEFworks/MUDAN documentation built on June 19, 2021, 6:46 a.m.