MultiK: MultiK main algorithm

In siyao-liu/MultiK: Objectively selects multiple insightful numbers of clusters K from single cell RNA-seq data.

Description

MultiK main algorithm: takes a preprocessed gene expression matrix as input. Then subsamples 80% of the cells and applies standard Seurat pipeline on the subsampled data matrix 100 times over multiple resolution parameters.

Usage

MultiK(
  seu,
  resolution = seq(0.05, 2, 0.05),
  nPC = 30,
  reps = 100,
  pSample = 0.8,
  seed = NULL
)

Arguments

seu	A Seurat object with normalized count
resolution	A vector Seurat resolution parameters. Default is from 0.05 to 2 with step size of 0.05
nPC	Number of principal components to use in clustering
reps	Number of subsampling runs. Integer value. Default is 100
pSample	Proportion of cells to sample. Numerical value. Default is 0.8
seed	Optional numerical value. This sets a random seed for generating reproducible results

Value

A list with components: k is a vector of number of runs for each K. clusters is a list containing the clustering labels for each subsampling run at each resolution parameter. consensus is a list containing a consensus matrix for each K.

siyao-liu/MultiK documentation built on Feb. 6, 2022, 1:42 a.m.