mosaicmpi.dataset.Dataset.get_category_overrepresentation

mosaicmpi.dataset.Dataset.get_category_overrepresentation#

Dataset.get_category_overrepresentation(layer: Collection[str] | str, truncate_negative: bool = False, subset_categories: Collection[str] = None) → DataFrame#

Calculate Pearson residual of chi-squared test, associating programs for each rank (k) to categories of samples/observations.

Parameters:

  • layer (Union[Collection[str], str]) – name of categorical data layer or combination of them

  • truncate_negative (bool, optional) – Truncate negative residuals to 0, defaults to False

  • subset_categories (Collection[str]) – Provide a subset of categories for calculating overrepresentation

Returns:

category × program matrix of overrepresentation values

Return type:

pd.DataFrame

Contents