Additional file 1 of Genome annotation with long RNA reads reveals new patterns of gene expression and improves single-cell analyses in an ant brain

Additional file 1: Figure S1. Statistics and methods used to create and evaluate the new Harpegnathos annotation. (A) Length distribution of all raw Iso-Seq subreads. (B) Transcripts per million (TPM) from short-read RNA-seq of genes with and without Iso-Seq coverage in brain and fat body/ovary. (C) Pipeline for manual annotation following combination of Iso-Seq-based and RNA-seq-based annotations. (D) Relationship between gene models in HSAL50 and HSAL51. (E) Schematic of the dT-seq approach. RNA was chemically fragmented. PolyA+ molecules were purified and split into two reverse transcription reactions, one primed with an anchored oligo-dT primer and one with random hexamers. The resulting cDNA was assembled into libraries and sequenced. The scheme at the bottom shows that the expected read distribution in dT- and hexamer-primed reactions differs for true polyA tails and internal A-stretches. This information was used to discard peaks that did not correspond to bona fide TTSs (red square). (F) Expected (top) and observed (bottom) signal at dT peaks found in the CDS (let) or 3′ UTR (right) from oligo-dT primed libraries (”dT”, top) and random hexamer primed libraries (”hexamers”, bottom).