mantispy.pp.filter_features

mantispy.pp.filter_features(adata, drop_nan=True, min_variance=0.0, blocklist='default', copy=False)

Drop all-NaN, low-variance and blocklisted features.

Parameters:

• adata (AnnData) – Object to filter.

• drop_nan (bool (default: True)) – Drop features that are missing everywhere.

• min_variance (float (default: 0.0)) – Drop features whose variance is at or below this, as feature_select() and sklearn’s VarianceThreshold do. 0 disables the check.

• blocklist (str | Sequence[str] | None (default: 'default')) – "default" for the bundled CellProfiler blocklist, an explicit list of names, or None to skip. Matched against the current names and against var["original_name"], so it works either side of standardize_feature_names().

• copy (bool (default: False)) – Return a filtered copy instead of filtering in place.

Return type:

AnnData | None

Returns:

None, or the filtered copy. Subsets var to the surviving features, and reports how many were dropped.