Hi, not really a bug, more of a documentation issue: sc.tl.rank_genes_groups seems to expect log-transformed data (be it in .X or .raw.X). To my knowledge this is not mentioned in the docs.I just ran ...

PandaPy has the speed of NumPy and the usability of Pandas 10x to 50x faster (by @firmai) - GitHub ... It is a fascinating Python library with a lot of potential to become mainstream." SSRN Report.