mapqc.evaluate

Contents

mapqc.evaluate#

mapqc.evaluate(adata, case_control_key, case_cats, control_cats)#

Evaluate and summarize mapQC output.

Parameters:

  • adata (AnnData) – AnnData object with mapQC output (i.e. mapqc.run() has been run)

  • case_control_key (str) – Column name in adata.obs that contains the case-control information

  • case_cats (list[str]) – Unique case categories, i.e. the non-control categories to be evaluated for the query cells only, as a list (e.g. [“IPF”, “COPD”]). Each category will be evaluated separately. If only one category exists, still provide as a list (e.g. [“IPF”]).

  • control_cats (list[str]) – Unique control categories in adata.obs[case_control_key] for the query cells only, as a list (e.g. [“Control”, “Control_2”]). These will be evaluated as one group. The controls are considered to be the same as the reference used to map against.

Return type:

dict

Returns:

dict of statistics: Dictionary containing the following statistics:

  • perc_nhoods_passfloat

    Percentage of neighborhoods that passed filtering

  • perc_cells_sampledfloat

    Percentage of cells that were sampled

  • perc_sampled_cells_passfloat

    Percentage of sampled cells that passed filtering

  • perc_[control_cat]_cells_dist_to_reffloat

    Percentage of [control_cat] cells that passed filtering that were distant to the reference (mapQC score > 2)

  • perc_[case_cat]_cells_dist_to_reffloat

    Percentage of [case_cat] cells that passed filtering that were distant to the reference (mapQC score > 2), for each case_cat included in case_cats

Notes

In addition to returning these statistics as a dictionary, the function prints each statistic to the console as it is calculated.

Contents