Metadata-Version: 2.1
Name: acdc-py
Version: 1.0.1
Summary: A package to quickly identify unbiased graph-based clusterings via parameter optimization in Python
Home-page: https://github.com/califano-lab/acdc_py
Author: Alexander L.E. Wang & Luca Zanella & Alessandro Vasciaveo
Author-email: aw3436@cumc.columbia.edu
License: MIT
Requires-Python: >=3.8
Description-Content-Type: text/markdown
License-File: LICENSE
Requires-Dist: scipy
Requires-Dist: tqdm
Requires-Dist: scanpy
Requires-Dist: anndata
Requires-Dist: pandas
Requires-Dist: numpy
Requires-Dist: joblib
Requires-Dist: leidenalg
Requires-Dist: scikit-learn
Requires-Dist: matplotlib
Requires-Dist: seaborn
Requires-Dist: datetime

This Scanpy-compatible package provides the ability to iterative compute clusterings using graph-based on combinations of resolution and KNN. At each point in this search space, a metric, such as silhouette score, is computed. The solution from this search space that maximizes the metric results in an optimal unbiased clustering solution. This process is sped up by our subsampling and diffusion feature that provides a near-identical solution in a much shorter time.
