Nurmansyah, Agnelo Furtado, Robert J Henry
Wild rice species harbour extensive but largely uncharacterised transcriptional diversity that underpins key agronomic traits and represents a valuable resource for crop improvement. Here we integrate single-molecule long-read sequencing (Iso-Seq) with short-read RNA-seq to resolve transcriptomes of Australian Oryza rufipogon-like ( O. rufipogon Aus) and O. meridionalis , the closest and most phylogenetically distant AA-genome relatives of Asian domesticated rice, respectively. Iso-Seq reveals substantially greater transcriptome complexity than represented in current reference annotations, identifying 48,876 transcript isoforms in O. rufipogon Aus and 69,180 in O. meridionalis . More than 60% of annotated genes are captured, and over 40% exhibit alternative isoforms, compared with approximately 10% in existing annotations. Integration of Iso-Seq data substantially improves genome annotations by restoring untranslated regions, increasing gene and transcript models, and improving BUSCO completeness. We identify 2,021 novel genes in O. rufipogon Aus and 2,645 in O. meridionalis , resolve hundreds of mis-assembled fusion genes, and uncover thousands of previously unannotated isoforms, including species-specific protein-coding transcripts and long non-coding RNAs. Analysis of three grain-size regulators ( OsMADS1 , GS1 , and SGW5 ) reveals substantial isoform diversity in wild rice. Notably, functional OsMADS1 and GS1 isoforms previously identified in domesticated rice populations are also detected in wild species, demonstrating that multiple functional transcript variants exist prior to domestication. Iso-Seq further identifies O. rufipogon -specific splice variants and reduced SGW5 expression associated with its narrow-grain phenotype. Our findings uncover hidden transcriptional complexity in wild rice, substantially refine genome annotations, and provide valuable resources for functional transcript studies and precise isoform-level quantification.