Strengths
- Scales efficiently to datasets exceeding one million cells.
- Comprehensive functionality covering preprocessing, visualization, clustering, trajectory inference, and differential expression testing.
- Active maintenance with 94 releases and a large contributor community.
- Open-source under BSD-3-Clause license with no cost.
- Integration with anndata and related Scanpy ecosystem tools.
Limitations
- Requires installation of additional dependencies like leidenalg and python-igraph for full functionality.
- Development version requires cloning the repository and installing in editable mode.
- Documentation building and contribution workflows involve specific tools such as Hatch and submodules.