mantispy.metrics.batch_variance_explained

mantispy.metrics.batch_variance_explained(adata, keys, use_rep='X_pca')

pc_regression() for several covariates, stacked into one frame.

Parameters:

• adata (AnnData) – Object with the embedding to measure in.

• keys (Sequence[str]) – obs columns to score, one row of the result each. Any column works, numeric or categorical, not only the batch: a plate position, a cell count or a treatment label are all valid covariates.

• use_rep (str (default: 'X_pca')) – obsm key of the embedding.

Return type:

DataFrame

Returns:

A tidy frame holding one pc_regression row per entry of keys.

Raises:

KeyError – obsm holds nothing under use_rep.