mantispy.pp.decorr_threshold_sweep

mantispy.pp.decorr_threshold_sweep(adata, thresholds, scorers, *, key='selected', **feature_select_kwargs)

Score decorrelate at a range of thresholds so the smallest set that holds a metric can be chosen.

decorr_threshold trades size for signal, and the right setting depends on the screen and on the metric that matters, so there is no universal best. This runs feature_select() with decorrelate=True at each threshold on a copy, scores the selected object with each callable, and returns one row per threshold.

Parameters:

• adata (AnnData) – Object to select features on. Never modified.

• thresholds (Sequence[float]) – decorr_threshold values to try.

• scorers (Mapping[str, Callable[[AnnData], float]] | Sequence[Callable[[AnnData], float]]) – Scoring callables, each taking the selected AnnData and returning a scalar (for example replicate or activity mAP). A mapping names the columns; a sequence names them by callable.__name__.

• key (str (default: 'selected')) – Boolean var column feature_select() writes and subset_features() reads.

• feature_select_kwargs (Any) – Passed through to feature_select() (operations, corr_threshold, and so on); decorrelate, decorr_threshold, key_added and copy are set here.

Return type:

DataFrame

Returns:

A frame with a threshold column, an n_kept column, and one column per scorer.