smart subsampling / sketching

What kind of feature would you like to request?

New analysis tool: A simple analysis tool you have been using and are missing in sc.tools?

Please describe your wishes

I'm currently implementing a function that takes in an anndata and then subsamples a given representation using https://github.com/jmschrei/apricot. This generally serves the purpose of semi-optimally picking a reduced number of points that's still representative of the latent space.

Is this sth within the scope of scanpy? When it's done it wouldn't be too much effort to polish it up for a PR. The dependency load seems fairly low.

methods

• Kernel Herding: https://github.com/CompCy-lab/SketchKH
• Geometric Sketching: https://github.com/brianhie/geosketch
• misc (scikit imbalanced-learn): https://imbalanced-learn.org/stable/index.html
• apricot: https://dl.acm.org/doi/pdf/10.1145/3388440.3412409

Start by reviewing the sc.tools analysis-tool area and the proposed anndata subsampling use case. Compare the listed approaches, including apricot, Kernel Herding, Geometric Sketching, and imbalanced-learn, before determining whether one belongs in Scanpy. Done would require an agreed scope and a polished PR for a representative latent-space subsampling tool.