background model / global pattern analysis

[soobeomlee]

Hi! Thanks for the very useful tool for single-cell APA analysis!

I have a few questions about the analysis.

First, how much does the selection of background cells matter when you have the mixture of disease cells and control cells with different cell types? For example, in my case, I am comparing leukemia cells and healthy cells where the data is biased to have more leukemia cells than healthy. In this case would you recommend to use control cells as background, or to subsample, or just use entire cells as background?

And I wonder if you can provide the script for the global shortening/lengthening analysis in figure3/4 in the paper. That would be super helpful.

Thanks!!

Best,
Soobeom