AdataDict Class

This module contains the AdataDict class, which is basically a nested dictionary of anndata, with a few extra features to help restructure the nesting hierarchy and iterate over it. AdataDict inherits from dict and passes methods through to each AnnData in the AdataDict.

class anndict.AdataDict(data=None, hierarchy=None)

AdataDict is a dictionary-like container where values are AnnData objects. AdataDict inherits from dict.

This class provides three main functionalities:

1. It has the set_hierarchy method to restructure the nesting hierarchy, and the hierarchy attribute to keep track.

2. It behaves like an AnnData object by passing methods through to each AnnData in the dictionary.

3. It has methods fapply(func, kwargs) and fapply_return(func, kwargs) that apply a given function func with arguments kwargs to each AnnData object in the AdataDict.

Parameters:

data dict[tuple[int, ...], any] | None (default: None)

Dictionary with keys as tuples of indices.

hierarchy tuple | list | None (default: None)

Tuple or list indicating the order of indices in the keys of data.

property hierarchy

The hierarchy of the AdataDict.

This attribute is accessed as: adata_dict.hierarchy.

Returns:

The current hierarchy of the AdataDict as a tuple.

Examples

adata_dict.hierarchy
> ('donor', ('tissue'))

Manipulate the hierarchy

AdataDict.set_hierarchy(nesting_list)	Rearrange the hierarchy of AdataDict based on the provided nesting structure.
AdataDict.add_stratification(strata_keys, *)	Split each value of an AnnData dictionary into further subsets based on additional desired strata.
AdataDict.flatten()	Flatten the hierarchy of AdataDict.

See also

add_stratification()

The function underneath add_stratification that can be used to return a new object instead of modifying in place.

Iterate over AdataDict

AdataDict.fapply(func, *[, ...])	Applies func to each AnnData in adata_dict, with error handling, retry mechanism, and the option to use either threading or sequential execution.
AdataDict.fapply_return(func, *[, ...])	Legacy wrapper for adata_dict_fapply.

See also

adata_dict_fapply()

The function underneath fapply that can be used separatley.

Index with a boolean mask

AdataDict.index_bool(index_dict[, inplace])

Index the AdataDict by a dictionary of indices.

See also

adata_dict_fapply()

Use this to generate a boolean mask of the correct format.

Set .obs and .var index

AdataDict.set_obs_index(column)	Set the index of adata.obs to the specified column for each AnnData object in adata_dict.
AdataDict.set_var_index(column)	Set the index of adata.var to the specified column for each AnnData in adata_dict.

Miscellaneous

AdataDict.check_structure(input_dict)	Check if an input_dict has the same structure and keys as the AdataDict.
AdataDict.copy()	Copy the AdataDict.