DE#

Differential expression results returned by Pseudobulk.de().

class brisc.DE(source=None, /, *, table=None, voom_weights=None, voom_plot_data=None)[source]#

Parameters:

source: str | Path | None
a directory containing a DE object saved with save(). Mutually exclusive with table, voom_weights, and voom_plot_data.
table: DataFrame | None
a polars DataFrame containing the DE results, with columns:
- cell_type: the cell type in which DE was tested
- coefficient: the coefficient (or contrast) for which DE was tested
- gene: the gene for which DE was tested
- logFC: the log2 fold change of the gene, i.e. its effect size
- SE: the standard error of the effect size
- LCI: the lower 95% confidence interval of the effect size
- UCI: the upper 95% confidence interval of the effect size
- AveExpr: the gene’s average expression in this cell type, in log CPM
- p: the DE p-value
- Bonferroni: the Bonferroni-corrected DE p-value
- FDR: the FDR q-value for the DE
Mutually exclusive with source.
voom_weights: dict[str, DataFrame] | None
an optional {cell_type: DataFrame} dictionary of voom weights, where rows are genes and columns are samples. The first column of each cell type’s DataFrame, ‘gene’, contains the gene names. Mutually exclusive with source.
voom_plot_data: dict[str, DataFrame] | None
an optional {cell_type: DataFrame} dictionary of info necessary to construct a voom plot with plot_voom(). Mutually exclusive with source.

Analysis#

`get_hits`	Get all (or the top) differentially expressed genes.
`get_num_hits`	Get the number of differentially expressed genes in each cell type.
`plot_voom`	Generate a voom plot for a cell type that differential expression was calculated for.
`plot_volcano`	Generate a volcano plot of DE hits.

Save a DE object to directory (which must not exist unless overwrite=True, and will be created) with the table at table.parquet.

a dictionary mapping cell type names to group names used by voomByGroup for that cell type, or None if voomByGroup was not used for that cell type.