tacco.utils

General utility functions

Functions

AnnData_query	Query the columns of .obs of an AnnData with a boolean expression and use the result to subset the AnnData.
anndata2dataframe	Creates a long form DataFrame from an AnnData.
bin	Bins points in position space.
cdist	Calclulate a dense pairwise distance matrix of sparse and dense inputs.
col_scale	Rescales columns of dense or sparse matrix inplace.
complete_choice	Similar to numpy.random.choice() with replace=False, but with special behaviour for size>a.shape[0].
coo_tocsr_buffered	Converts a sparse matrix in coo format into a sparse matrix in csr format consuming less memory than working in-memory by using hard disc buffer.
coo_tocsr_inplace	Converts a sparse matrix in coo format into a sparse matrix in csr format without allocating huge temporaries by reusing the memory of the input coo matrix.
cpu_count	Return the number of available CPU cores.
dataframe2anndata	Creates an AnnData from a long form DataFrame.
dense_distance_matrix	Calclulate a dense pairwise euclidean distance matrix of dense inputs.
divide	Calculates division out = a / b.
find_unused_key	Finds an unused key.
fishers_exact	Perform Fisher's exact test.
flat_inverse_mapping	Given a condensed mapping construct a flat inverse mapping.
gemmT	Perform a matrix-matrix multiplication A @ B.T for arbitrary sparseness of A and B in parallel.
generate_mixture_profiles	Creates symmetric pairwise mixture profiles.
get_average_profiles	Get average profiles for a certain annotation from a reference Anndata.
get_balanced_separated_intervals	Find the approximations of intervals with equal number of points in each interval which keep a specified minimum separation between them.
get_first_principal_axis	Get the first principal axis of a set of points.
get_maximum_annotation	Turns a soft annotation into a categorical annotation by reporting the annotation categories with the maximum values per oberservation.
get_sum	Calculates the sum of a sparse matrix or array-like in a specified axis and returns a flattened result.
hash	Create collision-free hash of several categorical columns by lexicograhical indexing.
heapsort3	Sorts an array in-place while following the reordering of the elements in two other arrays.
log	Calculates log inplace and in parallel.
log1p	Calculates log1p inplace and in parallel.
mannwhitneyu	Perform Mann-Whitney-U test.
merge_annotation	Merges annotation into coarser groups, e.g. from subtypes to types.
merge_colors	Merges a dict-like of colors into coarser colors.
min_dtype	Gives the smallest signed integer dtype which is necessary to represent maximum.
mode	Calculates the most frequent value per row in a dataframe.
parallel_nnls	Runs multiple nnls in parallel.
preprocess_single_cell_data	Preprocess single cell data in a standardized way from bare counts as input.
projection	Calculate pairwise normalized projections of sparse and dense inputs.
row_scale	Rescales rows of dense or sparse matrix inplace.
scale_counts	Scales the count matrix in an AnnData in various locations inplace.
solve_OT	Solve optimal transport problem with entropy regularization and optionally Kullback-Leibler divergence penalty terms instead of exact marginal conservation.
sparse_distance_matrix	Calclulate a sparse pairwise distance matrix of dense inputs.
sparse_result_gemmT	Perform a dense matrix-matrix multiplication A @ B.T for the case when only a sparse subset of the result is needed.
spatial_split	Splits a dataset into spatial patches balancing the number of observations per split.
split_spatial_samples	Splits a dataset into separated spatial patches.
sqrt	Calculates sqrt inplace and in parallel.
studentttest	Perform Student's t test.
tsne_single_cell_data	Preprocess single cell data in a standardized way from bare counts as input and include the generation of a tsne embedding.
umap_single_cell_data	Preprocess single cell data in a standardized way from bare counts as input and include the generation of a umap embedding.
welchttest	Perform Welch's t test.
write_adata_x_var_obs	Write only .X, .obs, and .var to an .h5ad file.