atac_mapper.topic_matching.TopicMatch

class atac_mapper.topic_matching.TopicMatch(region_topic)

A class to map a query dataset to the cistopics of a reference dataset.

Methods table

infer_topics(query[, njobs, n_iterations, tol])

Map a query dataset to the reference dataset

Methods

TopicMatch.infer_topics(query, njobs=-1, n_iterations=100, tol=0.0001)

Map a query dataset to the reference dataset

Args:

query: The query dataset in sparse matrix format (regions x cells matrix, must have the same

number of regions as the reference dataset).

njobs: Number of parallel jobs to run. -1 means using all available cores. n_iterations: Number of iterations for convergence. tol: Tolerance for convergence.

Return type:

ndarray

Returns:

topic_distributions: A numpy array of shape (n_cells, n_topics) containing the inferred topic distributions for each cell in the query dataset.