Preprocessing: PP#

Data integration#

pp.bbknn(adata, *[, batch_key, use_rep, ...])

Batch balanced kNN [Polański et al., 2019].

pp.harmony_integrate(adata, key, *[, basis, ...])

Use harmonypy [Korsunsky et al., 2019] to integrate different experiments.

pp.mnn_correct(*datas[, var_index, ...])

Correct batch effects by matching mutual nearest neighbors [Haghverdi et al., 2018] [Kang, 2018].

pp.scanorama_integrate(adata, key, *[, ...])

Use Scanorama [Hie et al., 2019] to integrate different experiments.

Sample demultiplexing#

pp.hashsolo(adata, cell_hashing_columns, *)

Probabilistic demultiplexing of cell hashing data using HashSolo [Bernstein et al., 2020].

Imputation#

Note that the fundamental limitations of imputation are still under debate.

pp.dca(adata[, mode, ae_type, ...])

Deep count autoencoder [Eraslan et al., 2019].

pp.magic(adata[, name_list, knn, decay, ...])

Markov Affinity-based Graph Imputation of Cells (MAGIC) API [van Dijk et al., 2018].