Issue with group and dedup option

[LioLenr]

Hello !

I am trying to detect viral transcript in RNA sec data (obtained with BD Rhapsody Enhanced bead 3').
In order to do that I am using UMI-tools to deal with the UMI and cell barcode and star aligner for the alignement.
My plan is to :
1 - Create a whitelist from the raw read using umi-tools -whitelist and then use umi-tools extract to clean my reads
2 - Create two index, one for the host and one for the virus with STAR and then align my processed reads with those index separately, again using STAR.
3 - Group and deduplicate the BAM files generated from step 2 with umi-tools group and dedup
4 - Generate two count matrix with umi-tools count with this shape cell_id X Host genes/Viral genes
5 - Merge the two count matrix using the cell_id as "anchor" point.

That is the plan.

Nevertheless, I am struggling to group and dedup the BAM files generated from the alignement of the reads with the viral index, not for the host index. I suppose that the really small size of the viral genome and index compared to the read size could be an issue.

I wanted to know if you can see other potentials issues and how to resolve them ?
Furthermore if you have done something related to what I am trying to do could you suggest me some other tool, approach or change to my approach ?

Best regards,

Lionel Lenoir