immuneML.pairwise_repertoire_comparison package

Submodules

immuneML.pairwise_repertoire_comparison.ComparisonData module

class immuneML.pairwise_repertoire_comparison.ComparisonData.ComparisonData(**kwargs)

Bases: object

add_items_for_repertoire(items: list, repertoire_id: str)

build_matching_fn()

filter_existing_items(items: list, repertoire_id: str) → list

get_batch(index: int, columns: Optional[list] = None, return_dict: bool = False)

get_batches(columns: Optional[list] = None, return_dict: bool = False)

get_item_names()

get_item_vector(index: int)

get_item_vectors(repertoire_ids: Optional[list] = None)

get_repertoire_vector(identifier: str)

get_repertoire_vectors(identifiers: list)

get_tmp_batches()

load_tmp_batch(batch_index: int) → dict

merge_tmp_batches_to_matrix()

process_dataset(**kwargs)

process_repertoire(**kwargs)

store_tmp_batch(batch: dict, batch_index: int)

immuneML.pairwise_repertoire_comparison.ComparisonDataBatch module

class immuneML.pairwise_repertoire_comparison.ComparisonDataBatch.ComparisonDataBatch(items: list, repertoire_index_mapping: Dict[str, int], path: pathlib.Path, identifier: int, matrix: Optional[numpy.ndarray] = None)

Bases: object

Parameters

• matrix – array with dimension items x repertoires, where items are defined by comparison attributes specified in ComparisonData class and can include, for instance, receptor sequences or combinations of receptor sequences and V and J gene

• items – the item names extracted from the repertoires in the dataset on which the repertoires are evaluated (e.g. sequences or combinations of sequences and genes

• repertoire_index_mapping – a mapping between the repertoire identifier (a string) and a column number for faster access of columns (repertoire vectors w.r.t. given items) in the comparison data matrix where columns correspond to repertoires

• path (Path) – path to directory where comp data is stored

• identifier (int) – identifier of the batch

get_items()

get_matrix()

identifier: int

items: list

load()

matrix: numpy.ndarray = None

path: pathlib.Path

repertoire_index_mapping: Dict[str, int]

store()

immuneML.pairwise_repertoire_comparison.PairwiseRepertoireComparison module

class immuneML.pairwise_repertoire_comparison.PairwiseRepertoireComparison.PairwiseRepertoireComparison(**kwargs)

Bases: object

compare(dataset: immuneML.data_model.dataset.RepertoireDataset.RepertoireDataset, comparison_fn, comparison_fn_name)

compare_repertoires(**kwargs)

create_comparison_data(**kwargs)

memo_by_params(dataset: immuneML.data_model.dataset.RepertoireDataset.RepertoireDataset)

prepare_caching_params(dataset: immuneML.data_model.dataset.RepertoireDataset.RepertoireDataset)

prepare_paralellization_arguments(repertoire_count: int, repertoire_identifiers: list, comparison_result)

Module contents