Tools: tl

Any transformation of the data matrix that is not preprocessing. In contrast to a preprocessing function, a tool usually adds an easily interpretable annotation to the data matrix, which can then be visualized with a corresponding plotting function.

Embeddings

tl.spectral(adata[, n_comps, features, ...])	Perform dimension reduction using Laplacian Eigenmaps.
tl.multi_spectral(adatas[, n_comps, ...])	Compute Laplacian Eigenmaps simultaneously on multiple modalities, with linear space and time complexity.
tl.umap(adata[, n_comps, use_dims, use_rep, ...])

Clustering

tl.leiden(adata[, resolution, ...])	Cluster cells into subgroups [Traag18].
tl.kmeans(adata, n_clusters[, n_iterations, ...])	Cluster cells into subgroups using the K-means algorithm, a classical algorithm in data mining.
tl.dbscan(adata[, eps, min_samples, ...])	Cluster cells into subgroups using the DBSCAN algorithm.
tl.hdbscan(adata[, min_cluster_size, ...])	Cluster cells into subgroups using the HDBSCAN algorithm.

Peak calling

tl.macs3(adata, *[, groupby, qvalue, ...])	Call peaks using MACS3.
tl.merge_peaks(peaks, chrom_sizes[, half_width])	Merge peaks from different groups.

Differential analysis

tl.marker_regions(data, groupby[, pvalue])	A quick-and-dirty way to get marker regions.
tl.diff_test(data, cell_group1, cell_group2)	Identify differentially accessible regions.

Motif analysis

tl.motif_enrichment(motifs, regions, ...[, ...])

Identify enriched transcription factor motifs.

Network analysis (beta)

tl.init_network_from_annotation(regions, ...)	Build CRE-gene network from gene annotations.
tl.add_cor_scores(network, *[, gene_mat, ...])	Compute correlation scores for any two connected nodes in the network.
tl.add_regr_scores(network, *[, peak_mat, ...])	Perform regression analysis for nodes and their parents in the network.
tl.add_tf_binding(network, *, motifs, ...[, ...])	Add TF motif binding information.
tl.link_tf_to_gene(network)	Contruct a genetic network by linking TFs to target genes.
tl.prune_network(network[, node_filter, ...])	Prune the network.

Utilities

tl.aggregate_X(adata[, groupby, normalize, file])	Aggregate values in adata.X in a row-wise fashion.
tl.aggregate_cells(adata[, use_rep, ...])	Aggregate cells into pseudo-cells.