It is very weird. I mapped a long-read fastq file with high quality isoforms to a reference genome, and got more mapped reads in the resulting .sam file than there were in the initial fastq file. I used parameters recommended by Cupcake: “-ax splice -t 30 -uf --secondary=no -C5” (https://github.com/Magdoll/cDNA_Cupcake/wiki/Cupcake:-supporting-scripts-for-Iso-Seq-after-clustering-step) The fastq file contained 44695 transcripts (which I counted by grep wc , and this number looks reasonable), and the mapped .sam file contains 46920 transcripts. I triple checked it.
Did someone experience something like this?