What is 'label' and 'replication' in my single-cell datasets?
Opened this issue · 1 comments
victorwang123 commented
Hi, my scRNA-seq datasets (from 10X genomics) clustered by seurat, and now I want to know the DEGs (differential expressed genes) in each subtypes (clusters), how to use Libra?
just DE = run_de(sc) seems wrong
jordansquair commented
I believe you are trying to find marker genes for each cluster? We did not discuss this in the paper, nor is Libra technically setup for one vs. all approaches.
You could consider a standard method for this (there are many, Seurat, scanpy, etc.). Alternatively you would need to manually setup a one vs. all loop. But I don't believe this has to do with a bug in Libra.