Methods for sorted cell populations

[li-wending]

Dear developers,

Thanks for this awesome tool!

I am trying to build the signature matrix from microRNA sequencing data from sorted cell populations, but the tutorial was only for single-cell inputs.

I wonder if there is any tutorial, or general guidance, for implementing AutoGeneS with sorted cell data?
Because for each cell type there are only a few (3~10) samples, and in my case there are only ~1k microRNAs available for analysis. Should I preselect highly variable genes as suggested for the single-cell data before building the Signature matrix?

Thanks in advance for your help!

Best,
Wending