rapids_singlecell icon indicating copy to clipboard operation
rapids_singlecell copied to clipboard

Feature: Support infercnvpy

Open xflicsu opened this issue 1 year ago • 2 comments

Hello, I try to perform CNV detection with infercnvpy with rapids_singlecell integrated 1 M cells. It still work slow. Any suggestion to speed up the process of CNV detection? Could rapids_singlecell add the CNV detection part (with GPU)? Thanks!

xflicsu avatar Jan 09 '24 03:01 xflicsu

Hey @xflicsu

Thanks for bringing up the speed issue with infercnvpy and rapids_singlecell for CNV detection in large datasets.

Integrating GPU support is an interesting idea. I'm considering it, but I'll need to see more interest from the user community to prioritize this development. If there’s enough demand, it's definitely on the table.

Also, any contributions to this effort are highly welcome. It would be great to collaborate on enhancing infercnvpy with GPU capabilities.

Let me know if you have further thoughts or if you're interested in contributing.

Severin

Intron7 avatar Jan 09 '24 09:01 Intron7

Just noticed this issue and wanted wanted to give a shoutout – I think it'd be awesome to have this feature. I feel that CNV and related analysis (prediction of aneuploidy, annotation of malignant cells...) seem to receive stronger attention in the R community (infercnv, copyKAT, SCEVAN, scATOMIC). Also, it looks to me that even though infercnvpy is lacking formal validation, it is gaining some credit, so this could be a good occasion to run some benchmarks and tests on it.

lorenzoamir avatar Feb 20 '24 15:02 lorenzoamir