Transcript Counting

Within each barcode and gene combination, IMIs are grouped in one of 64 bins, based on the 3-base binning index. For each bin, all identical IMIs are collapsed into a single count, since they are likely PCR duplicates of the same fragment generated during library prep.

Any barcode and gene combination that has ten or fewer unique binning indexes is assigned the number of unique binning indexes as its final count estimate. The pipeline then totals the number of IMIs associated with each remaining barcode and gene combination, and divides that number by the IPM correction factor, which accounts for the additional copies generated from a single captured molecule during five amplification cycles. The final count is the maximum between the floor of this value and the number of unique binning indexes for this barcode and gene.

For more information about Transcript Counting refer to the DRAGEN documentation detailing the DRAGEN PIPseq scRNA Pipeline.