sctools is a set of convenience functions for dealing with scRNAseq data within scanpy.
Functions to do fairly standard processing of scRNAseq data contained in scanpys AnnData objects. These functions essentially turn raw count data into normalized, clustered and dimension reduced data.
from sctools.pipeline import standard_processing
adata = standard_processing(adata)
sc.pl.umap(adata, color=['leiden', 'doublet_score'])
# differnetially expressed genes between clusters
sc.pl.rank_genes_groups(adata)
scanpy_DE_to_dataframe_fast()
turns scanpys awkward way of storing differential expression
into a bunch of pandas.DataFrame
s