Skip to contents

Simulate DE by swapping features

Source: R/simulateDE.R
simulateDE.Rd

A simple DE simulator that takes a dataset with a grouping factor and induces DE by swapping some features in one of the groups but not the other. This should ensure keeping the same data structure as the original data without having to estimate distribution parameters.

Usage

simulateDE(x, which_cols, ...)

# S4 method for ANY
simulateDE(x, which_cols, prop_DE = 0.01)

# S4 method for SummarizedExperiment
simulateDE(x, which_cols, ..., use_assay = "counts")

# S4 method for SingleCellExperiment
simulateDE(x, which_cols, ...)

Arguments

x

A numeric matrix containing features in rows and cells in columns. Alternatively, a SummarizedExperiment or SingleCellExperiment object.

which_cols

Character, numeric or logical vector specifying for which columns (cells, samples, ...) of x the feature swapping should occur. Usually, this will be a random subset of columns belonging to a mock group.

...

For the generic, arguments to be passed to specific methods.

For the SummarizedExperiment method, further arguments to be passed to the ANY method.

For the SingleCellExperiment method, further arguments to be passed to the SummarizedExperiment method.

prop_DE

Numeric scalar specifying the proportion of features that will be simulated as DE. Default: 0.01.

use_assay

A string or integer scalar specifying the assay of x to be used as input for the simulation. The default is to use "counts".

Value

A SummarizedExperiment object with DE induced between the specified groups. The simulated counts are in the "counts" assay of the returned object. The "sim_group" column of the colData indicates whether swapping was performed in that column (specified by the which_cols

argument). The rowData contains the following columns:

  • "is_DE": logical vector indicating the ground truth status of each feature

  • "swapped_feature": character vector indicating the original feature with which the feature was swapped

If x was a SummarizedExperiment object, the original colData and rowData are combined with these new columns.

If x was a SingleCellExperiment object, the output will also be a SingleCellExperiment.

Details

Note that for SummarizedExperiment objects, any additional assays (containing for example normalized values) are not retained. For simple library size-factor normalization, size factors for each cell will be identical before and after simulation. However, for more sophisticated methods, such as normalization by deconvolution, the size factors might differ due to the feature swapping. The expected number of DE features will be equal to prop_DE * nrow(x). Note however that it's possible that the actual number might be 1 lower than this. This can happen when a single feature remains un-"swapped" without any other features left to swap it with.

Author

Milan Malfait

Examples

example_sce <- scuttle::mockSCE()

## Swap features in the "treat1" group
cells_to_swap <- example_sce$Treatment == "treat1"

sim <- simulateDE(example_sce, which_cols = cells_to_swap, prop_DE = 0.1)
sim
#> class: SingleCellExperiment 
#> dim: 2000 200 
#> metadata(0):
#> assays(1): counts
#> rownames(2000): Gene_0001 Gene_0002 ... Gene_1999 Gene_2000
#> rowData names(2): is_DE swapped_feature
#> colnames(200): Cell_001 Cell_002 ... Cell_199 Cell_200
#> colData names(4): Mutation_Status Cell_Cycle Treatment sim_group
#> reducedDimNames(0):
#> mainExpName: NULL
#> altExpNames(0):