simComplexCounts: Simulate a complex counts matrix including different types of...

In zh542370159/dropSplit: Automatically identify cell-containing and empty droplets for droplet-based scRNAseq data using dropSplit

Simulate a complex counts matrix including different types of empty and cell-containing droplets.

Description

Simulation of a complex single-cell sequencing dataset.

Usage

simComplexCounts(
  total_gene = 30000,
  disturbance = 0.2,
  nempty = 50000,
  nlarge = 5000,
  nsmall = 500,
  empty_type = 2,
  large_type = 10,
  small_type = 2,
  empty_prof = NULL,
  empty_ngene_rate = 0.05,
  empty_rate = 1/100,
  large_prof = NULL,
  large_ngene_rate = 0.8,
  large_shape = 6,
  large_scale = 500,
  small_prof = NULL,
  small_ngene_rate = 0.6,
  small_shape = 10,
  small_scale = 100,
  large_frag = TRUE,
  large_frag_gene = 1:50,
  large_frag_prop = 0.5,
  small_frag = TRUE,
  small_frag_gene = 1:50,
  small_frag_prop = 0.5,
  remove_zero_drop = TRUE,
  remove_zero_feature = TRUE,
  seed = 0
)

Arguments

total_gene	Total gene number for all the simulated counts.
disturbance	A numeric value used as a weight of standard deviations when sample different distribution parameters for each cell/empty type from defined global parameters. Default is 0.2.
nempty, nlarge, nsmall	Empty, large cell and small cell droplets number. If remove_zero_drop is TRUE, the final number may be smaller beacuse droplets that have all zero-valued counts will be removed.
empty_type, large_type, small_type	Total number of types for nempty,nlarge,nsmall.
empty_prof, large_prof, small_prof	The overall gene expression profile distribution. If provided, must be the same length with total_gene.
empty_ngene_rate, large_ngene_rate, small_ngene_rate	Rate of total genes expressed in each type of droplets.
empty_rate	Rate parameter of exponential distribution for 'Empty'.
large_shape, small_shape	shape parameters in the GammaDist. shape*scale control the expected mean value of counts in large or small cells.
large_scale, small_scale	scale parameters in the GammaDist. shape*scale^2 control the expected variance of counts in large or small cells.
large_frag, small_frag	Whether simulate cell fragments from the large or small cells. Default is TRUE.
large_frag_gene, small_frag_gene	Indices of cell fragment gene in profile. Default is 1:100.
large_frag_prop, small_frag_prop	Proportion of the cell fragment gene counts. Default is 0.5.
remove_zero_drop	Whether to remove all zero-valued droplets.
remove_zero_feature	Whether to remove all zero-valued features.
seed	Random seed used in simulation. Default is 0.

Value

A sparse Matrix of class "dgCMatrix".

Examples

counts <- simComplexCounts()
counts

zh542370159/dropSplit documentation built on June 19, 2022, 2:49 p.m.