In a typical “barnyard” experiment in which cells from different species are mixed before loading to the 10x controller, the identification of the species of origin after mapping/counting with the hybrid reference is a problem. People tend to use the ratio of reads mapped to each reference genome to determine which species a cell is from.
In this paper https://www.biorxiv.org/content/10.1101/630087v1.full
To deconvolute species, detect doublets and low quality cells, the mixed-species …