sccoda.model.other_models.CLRModel_ttest.eval_model

CLRModel_ttest.eval_model(ground_truth, alpha=0.05, fdr_correct=True)

Evaluates array of p-values compared to a ground truth via binary classification.

Parameters

ground_truth : ListList

List (boolean, length same as number of cell types) indicating differential abundance for each cell type

alpha : floatfloat (default: 0.05)

p-value (or q-value if using FDR correction) threshold

fdr_correct : boolbool (default: True)

Whether to use Benjamini-Hochberg FDR correction for multiple testing

Return type

Tuple[ndarray, ndarray, ndarray, ndarray]Tuple[ndarray, ndarray, ndarray, ndarray]

Returns

• classification results

• tp, tn, fp, fn – Number of True positive, … effects