structboost.sim_scrnaseq_anndata

structboost.sim_scrnaseq_anndata(*, n=1000, n_genes=50, stageno=10, stagep=None, stagen=None, stageoverlap=None, hierarchy=True, markers_per_level=None, imbalanced=False, base_mean=2.0, gene_mean_shape=4.0, effect_size=10.0, effect_size_sd=0.0, dispersion=0.2, lib_size_sd=0.0, n_batches=1, batch_effect_sd=0.0, ambient_frac=0.0, dropout_mid=None, dropout_shape=-1.0, seed=1, target_sum=10000.0, standardize=True)

Simulate scRNA-seq counts and package them as an AnnData object.

Wraps sim_scrnaseq_data() and applies the standard preprocessing pipeline. Raw counts and library-size-normalized log1p values are always kept as layers, so adata.X can be swapped without re-simulating.

Only the two preprocessing arguments are documented below; every other parameter is passed through to sim_scrnaseq_data() unchanged.

Parameters:

• target_sum (float) – Library size each cell is normalized to before log1p.

• standardize (bool) – If True, adata.X holds the per-gene z-score of the log1p-normalized values, ready for BAE. If False, adata.X holds the log1p-normalized values themselves.

• n (int)

• n_genes (int)

• stageno (int)

• stagep (int | None)

• stagen (int | None)

• stageoverlap (int | None)

• hierarchy (tuple[int, ...] | bool | None)

• markers_per_level (tuple[int, ...] | None)

• imbalanced (bool)

• base_mean (float)

• gene_mean_shape (float)

• effect_size (float)

• effect_size_sd (float)

• dispersion (float)

• lib_size_sd (float)

• n_batches (int)

• batch_effect_sd (float)

• ambient_frac (float)

• dropout_mid (float | None)

• dropout_shape (float)

• seed (int)

Returns:

anndata.AnnData – Shape (n, n_genes) with:

• X — z-scored log1p values, or log1p values when standardize=False

• layers["counts"] — raw integer UMI counts

• layers["lognorm"] — log1p of library-size-normalized counts

• obs["stage"] — ground-truth stage label per cell

• obs["stage_id"], obs["batch"], obs["size_factor"], obs["total_counts"]

• var["is_marker"], var["marker_stages"], var["n_marker_stages"], var["base_mean"]

• varm["marker_mask"] — (n_genes, stageno) ground-truth marker matrix

• uns["simulation"] — simulation parameters

Raises:

• ImportError – If anndata is not installed.

• ValueError – If target_sum <= 0 or any parameter of sim_scrnaseq_data() is out of range.

Return type:

ad.AnnData

Examples

>>> from structboost import sim_scrnaseq_anndata
>>> adata = sim_scrnaseq_anndata(n=100, n_genes=20, stageno=4, seed=0)
>>> adata.obs["stage"].nunique()
4