Print a specified warning to the screen, using either warnings or scanpy. Input ----- message : ``str`` The warning message to print scanpy_

Checks if the provided metric can be used with the implied KNN algorithm. Returns parameters with the metric validated. Input ----- params : `

Copied out of scanpy.neighbors This is from umap.fuzzy_simplicial_set [McInnes18]_. Given a set of data X, a neighborhood size, and a measure of

Create a faiss/cKDTree/KDTree/annoy/pynndescent index for nearest neighbour lookup. All undescribed input as in ``bbknn.bbknn()``. Returns the res

Identify the KNN structure to be used in graph construction. All input as in ``bbknn.bbknn()`` and ``bbknn.matrix.bbknn()``. Returns a tuple of dis

Copied out of scanpy.neighbors

Do pre-1.6.0 computation algorithm selection based on possible legacy arguments. Looks at the following (default None) parameters: approx, use_ann

Query the faiss/cKDTree/KDTree/annoy index with PCA coordinates from a batch. All undescribed input as in ``bbknn.bbknn()``. Returns a tuple of dis

Trims the graph to the top connectivities for each cell. All undescribed input as in ``bbknn.bbknn()``. Input ----- cnts : ``CSR`` Sparse mat

Scanpy-independent BBKNN variant that runs on a PCA matrix and list of per-cell batch assignments instead of an AnnData object. Non-data-entry argu

Batch balanced KNN, altering the KNN procedure to identify each cell's top neighbours in each batch separately instead of the entire cell pool with

Helper post-processing function that extracts a single cell's connectivity and stores it in ``adata.obs``, ready for plotting. Connectivities range

Perform ridge regression on scaled expression data, accepting both technical and biological categorical variables. The effect of the technical var