Hello all, I am trying to generate a count matrix with intronic and exonic abundances to use in a single-cell velocity analysis. I would like to start from my 10x BAM file (rather than the fastq reads) and I am considering dropEst and velocyto.

However, I am running into a problem - we combined two sequencing runs together into one BAM, and so our BAM file has barcode duplicates - as an example, my file might contain both 'CAAGACTTCTCGGCTT' and 'CAAGACTTCTCGGCTT-1'.

What is the best way to deal with having non-standard barcode format in my BAM?

Would I simply add all unique barcodes in my BAM to barcodes.tsv? Does either dropEst or velocyto also require that all barcodes be just 16 characters?

Thanks for any help/advice you can offer.