Annotatable

class openomics.database.base.Annotatable

Bases: abc.ABC

This abstract class provides an interface for the -omics (Expression) to annotate its genes list with the external data downloaded from various databases. The database will be imported as attributes information to the genes’s annotations, or interactions between the genes.

Attributes Summary

DISEASE_ASSOCIATIONS_COL
SEQUENCE_COL_NAME

Methods Summary

annotate_attributes(database, on, columns[, …])	Performs a left outer join between the annotation and Database’s DataFrame, on the index key.
annotate_diseases(database, index)	param database
annotate_expressions(database, index[, …])	Annotate :param database: :param index: :param fuzzy_match:
annotate_interactions(database, index)	param database
annotate_sequences(database, index[, agg, omic])	Annotate a genes list (based on index) with a dictionary of <gene_name: sequence>.
get_annotation_expressions()
get_annotations()
get_rename_dict(from_index, to_index)	Used to retrieve a lookup dictionary to convert from one index to another, e.g., gene_id to gene_name, obtained from two columns in the data frame.
initialize_annotations(index[, gene_list])	param index
set_index(new_index)	Resets :param new_index: :type new_index: str

Attributes Documentation

DISEASE_ASSOCIATIONS_COL = 'disease_associations'

SEQUENCE_COL_NAME = 'sequence'

Methods Documentation

annotate_attributes(database, on, columns, agg='concat', fuzzy_match=False)

Performs a left outer join between the annotation and Database’s DataFrame, on the index key. The index argument must be column present in both DataFrames. If there exists overlapping columns from the join, then .fillna() is used to fill NaN values in the old column with non-NaN values from the new column.

Parameters

• database (Database) – Database which contains an dataframe.

• on (str) – The column name which exists in both the annotations and Database dataframe to perform the join on.

• columns ([str]) – a list of column name to join to the annotation.

• agg (str) – Function to aggregate when there is more than one values for each index instance. E.g. [‘first’, ‘last’, ‘sum’, ‘mean’, ‘concat’], default ‘concat’.

• fuzzy_match (bool) – default False. Whether to join the annotation by applying a fuzzy match on the index with difflib.get_close_matches(). It is very computationally expensive and thus should only be used sparingly.

annotate_diseases(database, index)

Parameters

• database (DiseaseAssociation) –

• index (str) –

annotate_expressions(database, index, fuzzy_match=False)

Annotate :param database: :param index: :param fuzzy_match:

Parameters

• database –

• index –

• fuzzy_match –

annotate_interactions(database, index)

Parameters

• database (Interactions) –

• index (str) –

annotate_sequences(database, index, agg='longest', omic=None, **kwargs)

Annotate a genes list (based on index) with a dictionary of <gene_name: sequence>. If multiple sequences per gene name, then perform some aggregation.

Parameters

• database (Database) – The database

• index (str) – The gene index column name.

• agg (str) – The aggregation method, one of [“longest”, “shortest”, or “all”]. Default longest.

• omic (str) – Default None. Declare the omic type to fetch sequences for.

• **kwargs –

get_annotation_expressions()

get_annotations()

get_rename_dict(from_index, to_index)

Used to retrieve a lookup dictionary to convert from one index to another, e.g., gene_id to gene_name, obtained from two columns in the data frame.

Returns

Dict[str, str]: the lookup dictionary.

Parameters

• from_index (str) – an index on the DataFrame for key

• to_index –

initialize_annotations(index, gene_list=None)

Parameters

• index –

• gene_list –

set_index(new_index)

Resets :param new_index: :type new_index: str

Parameters

new_index –