Can we speed up run_de with mixed models ?

[Nayrouz109]

Hello,

Quick question, is it normal for run_DE(sc, de_family = "mixedmodel") to take too long to run (around 24hrs) when using one of your 18 gold standard single-cell datasets.

If so, is there a way you'd recommend using to speed it up?

Thanks!

[skinnider]

Short answer is this is normal and we did not find a way to speed it up using this particular implementation. You might be interested in NEBULA (https://www.nature.com/articles/s42003-021-02146-6) which seems like a promising alternative.

[Nayrouz109]

ah thank you !