pyproteonet.imputation.simple.across_sample_aggregate_impute

pyproteonet.imputation.simple.across_sample_aggregate_impute(dataset: Dataset, molecule: str, column: str, method: Literal['mean', 'median'] = 'mean', all_missing_percentile: float | None = None, all_missing_constant: float = 0)

Aggregates across samples to impute missing values, if all values are missing for a given molecule, then imputes using a low percentile value.

Parameters:

  • dataset (Dataset) – Dataset to imupute.

  • molecule (str) – Molecule type to impute (e.g. protein, peptide etc.).

  • column (str) – Value column with missing values to impute.

  • method (Literal['mean', 'median'], optional) – Across sample aggregation method. Defaults to ‘mean’.

  • all_missing_percentile (Optional[float], optional) – Percentile to use when all value are missing for a molecule. If not given a constant value is used. Defaults to None.

  • all_missing_constant (float, optional) – Constant value to use when molecule is missing across all samples and no all missing percentile is given. Defaults to 0.

Raises:

AttributeError – _description_

Returns:

_description_

Return type:

_type_