The high error using HISAT2 on separate genomes led us to consider using Kallisto, which uses pseudoalignment to efficiently quantify abundances of transcripts.44 Unlike HISAT2, which indexes the entire genome, Kallisto builds an index considering only subsequences, or k-mers, that distinguish isoforms. We built an index of the pooled mouse and human cDNA sequence, intending to focus on only the k-mers that distinguish the two different species (as well as different isoforms or genes within a species).
