c2v.utils.impute

c2v.utils.impute(adata, obs_name, value_to_impute='NA', use_rep='X_pca', weights=None, classification_obsm='impute_prob', key_added='imputed', gaus_sigma=None, k=10)

Impute missing values in an observation column using k-nearest neighbors.

Parameters:

adata sc.AnnData

AnnData object with observation values in obs[obs_name].

obs_name str

Name of the observation column to impute.

value_to_impute str, optional

Value in obs_name to impute. Default is “NA”.

use_rep str, optional

Key in obsm to use for neighbor search. Default is “X_pca”.

weights Literal["gaussian", "linear"] | None, optional

Weighting scheme for imputation. Default is None.

classification_obsm str, optional

Key in obsm to use for classification probabilities. Default is “impute_prob”.

key_added str, optional

Key in obs to store imputed values. Default is “imputed”.

gaus_sigma float | None, optional

Sigma for Gaussian weighting. If None, use median distance of positive neighbors. Default is None.

k int, optional

Number of neighbors to use for imputation. Default is 10.

Returns:

None Imputed values are stored in obs[f”{obs_name}_{key_added}”].