RunRobPCA: Perform robust PCA (Hubert PCA) with modified PC scores on a...

In gmstanle/scRobustPCA: Performs robust PCA with modified PCs on Seurat objects for scRNA-Seq dimensionality reduction

Description

scRobustPCA performs the variant of robust PCA developed by Mia Hubert et al. on the gene expression matrix "data" in a Seurat object. Set reduction.type='rpca' in other Seurat functions to use the rPCA results, for example to calculate clusters with FindClusters.

Usage

RunRobPCA(object, npcs = 10, pc.genes = NULL,
  use.modified.pcscores = TRUE)

Arguments

object	Seurat object
npcs	Number of principal components to calculate
pc.genes	Genes as input to PC. If pc.genes==NULL, the var.genes slot is used. If var.genes is empty and pc.genes==NULL, then all genes are used.
use.modified.pcscores	If FALSE, then the raw pc score output from robust PCA is used. If TRUE, then pc scores are replaced with the sum of the top 30 genes by positive loading minus the sum of the top 30 genes by negative loading. Each gene is scaled to a max of 1 and min of 0.

Value

A Seurat object with the 'rpca' field filled.

gmstanle/scRobustPCA documentation built on May 10, 2019, 10:01 a.m.