How can I use Azimuth with both scRNA-seq count matrix and ADT count matrix from in-house CITE-Seq data?

[qicheng-ma]

Dear Dr. Satija

"If uploading a Seurat object, it must contain an assay named ‘RNA’ with raw data in the ‘counts’ slot. Note that Azimuth uses only the (unnormalized) counts matrix."

Could you please tell me how I can use Azimuth with both scRNA-seq count matrix and ADT count matrix from in-house CITE-Seq data?

Thanks in advance!

[andrewwbutler]

Hi,

You can still use Azimuth with your CITE-seq dataset. It's just that the ADT information in the query will not be used in mapping. You can run the following to extract just the RNA data and then upload that reduced object to Azimuth for mapping.

DefaultAssay(object) <- “RNA”
object <- DietSeurat(object = object, assays = “RNA”)

[ravipatel4]

Hi @andrewwbutler,

Is there a plan to allow using our own CITE-seq dataset for mapping? Azimuth struggles distinguishing some cell types in our PBMC datasets, in particular, naïve vs memory T cell populations. We find that our CITE-seq data has a significant differentiating power and if there was a way to incorporate that into Azimuth, I would think the Azimuth annotation will improve quite a bit for the subsets of T cells.

Thanks.