Hi all,
I’m working with 10x multiome data and have processed the ATAC and RNA modalities separately:
For ATAC, I used tile matrix (e.g., 5kb bins) and trained a peakVI model.
For RNA, I used scVI.
Now, I’m planning to use MultiVI for integrative analysis. I have two options:
Option A: Use the clusters derived from peakVI to perform peak calling per cluster, merge the peaks, build a unified peak matrix across all cells, and then feed that into MultiVI.
Option B (preferred): Skip peak calling entirely, and directly use the tile matrix as input for MultiVI.
My questions are:
Does MultiVI officially support the use of tile matrix as input for ATAC modality?
Would the integration performance or downstream analyses (e.g., latent space, cell type separation) be significantly affected?
Any insights or recommendations from the devs or users who’ve tried both would be greatly appreciated!
Thanks!
Pam