cassiopeia.data.sample_bootstrap_allele_tables

cassiopeia.data.sample_bootstrap_allele_tables(allele_table, indel_priors=None, num_bootstraps=10, random_state=None, cut_sites=None)

Generates bootstrap character matrices from an allele table.

This function will take in an allele table, generated with the Cassiopeia preprocess pipeline and produce several bootstrap character matrices with respect to intBCs rather than individual cut-sites as in sample_bootstrap_character_matrices. This is useful because oftentimes there are dependencies between cut-sites on the same intBC TargetSite.

Parameters

allele_table : DataFrameDataFrame

AlleleTable from the Cassiopeia preprocessing pipeline

indel_priors : DataFrame, NoneOptional[DataFrame] (default: None)

A dataframe mapping indel identities to prior probabilities

num_bootstraps : intint (default: 10)

number of bootstrap samples to create

random_state : RandomState, NoneOptional[RandomState] (default: None)

A numpy random state for reproducibility.

cut_sites : List[str], NoneOptional[List[str]] (default: None)

Columns in the AlleleTable to treat as cut sites. If None, we assume that the cut-sites are denoted by columns of the form “r{int}” (e.g. “r1”)

Return type

List[Tuple[DataFrame, Dict[int, Dict[int, float]], Dict[int, Dict[int, str]], List[str]]]List[Tuple[DataFrame, Dict[int, Dict[int, float]], Dict[int, Dict[int, str]], List[str]]]

Returns

A list of bootstrap samples in the form of tuples

(bootstrapped character matrix, prior dictionary, state to indel mapping, bootstrapped intBC set)

cassiopeia.data.get_lca_characters cassiopeia.data.sample_bootstrap_character_matrices