batch effect removal for multi-sample AS data combination

[wangmeijiao]

Hi yang,
Here are two quick question: 1) if we want to merge multiple sample of AS single cell data, what strategy do you recommend to use to remove batch effect? 2) the RNA velocity analysis (scVelo, https://scvelo.readthedocs.io/en/stable/) can recapture the potential differential procedure with splicing/unsplicing ratio, however, the splicing data is hard to do batch effect removal. what do you suggest the best way to remove batch effect in velocity matrix?

Best,

Meijiao

[ssyshh]

Hello, your question is very interesting. I would like to ask if you successfully removed batch effects, and at which step did you remove them? How can I use the matrix after batch effect removal to perform UMAP dimensionality reduction and clustering?

[thestarocean]

I also have the same question here regarding the batch effect removal. According to our data, the batch effect seemed to be obvious in splicing landscape between samples.