Integrating Iso-seq and RNA-seq data for the reannotation of the killifish telencephalon transcriptome.

Abstract

The short-lived and rapidly aging African turquoise killifish, Nothobranchius furzeri GRZ is a unique model to study vertebrate aging. Current genomic and full-length transcriptomic sequencing lacks full gene annotations, resulting in poor mapping in bulk and single-cell transcriptomic studies. In our efforts to reannotate the transcriptome of the killifish telencephalon, we combined long-read (Smrt-Isoseq) and short-read transcriptome sequencing approaches. A total of 17,008 full-length isoforms, including 6763 novel ones were obtained (51 bp to 7,500 bp). The killifish telencephalon comprises 25% multi-exon genes, while over 50% are mono-exon genes. We discovered novel non-coding and coding sequences in both young and aged telencephali. We integrated long-read and RNA-seq data to construct a comprehensive transcriptome and profiled expression dynamics across the aging telencephalon. Our gene models demonstrate greater detail and accuracy than Ensembl, with more precise polyA locations. Alternative splicing analysis revealed 29 events altered with aging, which involved changes in ribosome function, gap junction and mRNA surveillance pathways. These generated resources pave the way for future functional genomic studies in this biogerontology model.