ruvSimulate: Simulate a simple matrix or SingleCellExperiment to test...

In SydneyBioX/scMerge: scMerge: Merging multiple batches of scRNA-seq data

Simulate a simple matrix or SingleCellExperiment to test internals of scMerge

Description

This function is designed to generate Poisson-random-variable data matrix to test on the internal algorithms of scMerge. It does not represent real biological situations and it is not intended to be used by end-users.

Usage

ruvSimulate(
  m = 100,
  n = 5000,
  nc = floor(n/2),
  nCelltypes = 3,
  nBatch = 2,
  k = 20,
  lambda = 0.1,
  sce = FALSE
)

Arguments

m	Number of observations
n	Number of features
nc	Number of negative controls
nCelltypes	Number of cell-types
nBatch	Number of batches
k	Number of unwanted factors in simulation
lambda	Rate parameter for random Poisson generation
sce	If TRUE, returns a SingleCellExperiment object

Value

If sce is FALSE, then the output is a list consists of

• Y, expression matrix generated through Poisson random variables,

• ctl, a logical vector indicating the control genes,

• M, replicate mapping matrix,

• cellTypes, a vector indicating simulated cell types

• batch, a vector indicating simulated batches

if sce is TRUE, a SingleCellExperiment wrapper will be applied on all above simulated objects.

Examples

set.seed(1)
L = ruvSimulate(m = 200, n = 1000, nc = 200, 
nCelltypes = 3, nBatch = 2, lambda = 0.1, k = 10, sce = TRUE)
print(L)
example <- scMerge(sce_combine = L,
                      ctl = paste0('gene', 1:500),
                      cell_type = L$cellTypes,
                      ruvK = 10,
                      assay_name = 'scMerge')
L = scater::runPCA(L, exprs_values = "logcounts")                      
scater::plotPCA(L, colour_by = 'cellTypes', shape = 'batch')

example = scater::runPCA(example, exprs_values = 'scMerge')                                       
scater::plotPCA(example, colour_by = 'cellTypes', shape = 'batch')

SydneyBioX/scMerge documentation built on April 14, 2024, 3:22 a.m.