Generates artificial pseudobulk samples from single-cell reference data and returns them in the genes x samples format used by the OmicsTweezer workflow.
Usage
omics_simulate(
sc_data,
d_prior = NULL,
n = 500L,
samplenum = 5000L,
random_state = NULL,
sparse = FALSE,
sparse_prob = 0.5,
rare = FALSE,
rare_percentage = 0.4,
celltype_col = "CellType",
assay = "RNA",
slot = "counts"
)Arguments
- sc_data
A single-cell reference object. Supported inputs are a `Seurat` object, a `SingleCellExperiment` object, a matrix, or a `data.frame`.
- d_prior
Optional numeric vector of Dirichlet concentration parameters. If `NULL`, a symmetric Dirichlet prior of ones is used.
- n
Integer scalar. Target number of single cells per simulated pseudobulk before rounding to integer cell counts.
- samplenum
Integer scalar. Number of pseudobulk samples to generate.
- random_state
Optional integer scalar used to control reproducible simulation.
- sparse
Logical scalar. If `TRUE`, a subset of simulated samples is forced to contain zero fractions for a subset of cell types.
- sparse_prob
Numeric scalar. Controls both the proportion of sparse samples and the proportion of cell types zeroed within those samples.
- rare
Logical scalar. If `TRUE`, a subset of cell types is perturbed to have very small fractions in a subset of samples.
- rare_percentage
Numeric scalar in `[0, 1]`. Fraction of cell types to treat as rare when `rare = TRUE`.
- celltype_col
Character scalar giving the metadata column that contains cell-type labels for `Seurat` and `SingleCellExperiment` input.
- assay
Character scalar giving the assay name to extract from a `Seurat` object.
- slot
Character scalar giving the assay slot or assay name to extract. For `Seurat`, this is interpreted as a slot or layer within `assay`. For `SingleCellExperiment`, this is interpreted as an assay name.
Value
A named list with components:
- X
Numeric matrix with genes in rows and simulated pseudobulk samples in columns.
- obs
Data frame with samples in rows and realized cell-type proportions in columns.
- var
Data frame indexed by gene names.
Details
Counts are sampled with replacement within each cell type according to Dirichlet-generated mixture proportions. Optional sparse and rare-cell perturbations can be applied to mimic heterogeneous cellular compositions.
The simulation follows the OmicsTweezer-style pseudobulk generation strategy. Single-cell counts are first grouped by cell type, then sampled according to Dirichlet-generated proportions and aggregated into pseudobulk profiles.