Count SE BC + UMI data

[zkliesmete]

Hi,
I was wondering whether for the counting part there is a possibility to use SE BC fastq (read1) and supply a UMI fastq (read2)?
In addition, a few things are unclear to me regarding the workflow:

1. is UMI collapsing happening per BC-UMI or across BCs per UMI? If the same UMI is associated with multiple BCs, how is this being handled?
2. Does it require a perfect match between the association file BC sequences and the ones observed in read1 fastq?
3. Related to 2), are the base qualities somehow taken into account?

Thank you a lot!