Wild rice (Oryza rufipogon) constitutes a primary gene source for rice breed improvement. Chromosome segment substitution line (CSSL) for O. rufipogon is a powerful tool for fine mapping of quantitative traits, new gene discovery, and marker-assisted breeding. Thus, they provide a basis for a wide range of genomic and genetic studies. In this study, a set of 198 CSSLs were developed from a cross between recurrent parent indica var. 9311 and an O. rufipogon donor parent; these were then genotyped using 313 polymorphic SSR markers evenly distributed across the 12 rice chromosomes. On average, each CSSL carried 2.16 introgressed segments, and the genetic distance of each segment was about 6 cM. The segments collectively covered 84.9 % of the wild rice genome. Based on these CSSLs, 25 QTLs involved in 10 agronomic traits were identified. Seven CSSLs were subjected to a whole-genome single nucleotide polymorphism chip assay and two QTLs, qSH4-1 and qDTH10-1, detected. In addition, a new QTL associated with the heading date was detected in a 78-Kb region on chromosome 10, thus proving the ability of these CSSLs to identify new QTLs and genes. The newly developed CSSL population proved a useful tool for both gene identification and whole-genome research of wild rice. These CSSL materials will provide a foundation for rice variety improvement.

Summary: Germplasm resources serve as the foundations of advancements in breeding and are crucial for maintaining food security. Wild rice species of the genus Oryza include rich sources of genetic diversity and high adaptability, making them a substantial resource for rice breeding. The discovery of wild‐type cytoplasmic male sterility resources enabled the achievement of the ‘three lines’ goal in hybrid rice, significantly increasing rice yields. The application of resistance alleles from wild rice enables rice production to withstand losses caused by stress. Reduced genetic diversity due to rice breeding poses a significant limitation to further advances and can be alleviated through a systematic use of wild genetic resources that integrate geographic, climatic and environmental data of the original habitat, along with extensive germplasm collection and identification using advanced methods. Leveraging technological advancements in plant genomics, the understanding of genetic mechanisms and the application of artificial intelligence and gene editing can further enhance the efficiency and accuracy of this process. These advancements facilitate rapid isolation and functional studies of genes, and precise genome manipulation. This review systematically summarizes the utilization of superior genes and germplasm resources derived from wild rice sources, while also exploring the collection, conservation, identification and utilization of further wild rice germplasm resources. A focus on genome sequencing and biotechnology developments is leading to new breeding and biotechnology opportunities. These new opportunities will not only promote the development of rice varieties that exhibit high yields, superior stress resistance and high quality but also expand the genetic diversity among rice cultivars.

Genomes carry millions of noncoding variants, and identifying the tiny fraction with functional consequences is a major challenge for genomics. We assessed the role of selection on long noncoding RNAs (lncRNAs) for domestication-related changes in rice grains. Among 3363 lncRNA transcripts identified in early developing panicles, 95% of those with differential expression (329 lncRNAs) between Oryza sativa ssp. japonica and wild rice were significantly down-regulated in the domestication event. Joint genome and transcriptome analyses reveal that directional selection on lncRNAs altered the expression of energy metabolism genes during domestication. Transgenic experiments and population analyses with three focal lncRNAs illustrate that selection on these loci led to increased starch content and grain weight. Together, our findings indicate that genome-wide selection for lncRNA down-regulation was an important mechanism for the emergence of rice domestication traits.

The development of germplasm resources and advances in breeding methods have led to steady increases in yield and quality of rice (Oryza sativa L.). Three milestones in the recent history of rice breeding have contributed to these increases: dwarf rice breeding, hybrid rice breeding, and super rice breeding. On the 50th anniversary of the success of three‐line hybrid rice, we highlight important scientific discoveries in rice breeding that were made by Chinese scientists and summarize the broader history of the field. We discuss the strategies that could be used in the future to optimize rice breeding further in the hope that China will continue to play a leading role in international rice breeding.

Crop breeding technologies are vital for global food security. While traditional methods have improved yield, stress tolerance, and nutrition, rising challenges such as climate instability, land loss, and pest pressure now demand new solutions. This study introduces the Breeding 5.0 framework, driven by artificial intelligence (AI) and robotics, marking a shift from empirical selection to intelligent systems. Central to this transformation is AI's emerging ability to deeply “understand germplasm,” not merely by identifying genetic markers but also by decoding its architecture, plasticity, regulatory logic, and environmental interactions. This germplasm intelligence enables predictive trait modeling, optimized parental design, and targeted selection. We define four technical paradigms enabling this shift: (i) Multimodal data integration to bridge genotype and phenotype; (ii) Omni‐simulated environments for virtual performance testing; (iii) Peopleless data capture for scalable precision; and (iv) Expert, explainable AI for biologically grounded decisions. Together, these technologies algorithmically convert germplasm into actionable breeding insights, accelerating the full cycle from ideal plant type design to elite line development. We further propose the “breeding flywheel,” a self‐reinforcing system that continuously amplifies phenotypic gains and refines breeding strategies, thereby enabling faster and smarter crop improvement to ensure a sustainable food future.

Rice (Oryza sativa) is a significant crop worldwide with a genome shaped by various evolutionary factors. Rice centromeres are crucial for chromosome segregation, and contain some unreported genes. Due to the diverse and complex centromere region, a comprehensive understanding of rice centromere structure and function at the population level is needed. We constructed a high‐quality centromere map based on the rice super pan‐genome consisting of a 251‐accession panel comprising both cultivated and wild species of Asian and African rice. We showed that rice centromeres have diverse satellite repeat CentO, which vary across chromosomes and subpopulations, reflecting their distinct evolutionary patterns. We also revealed that long terminal repeats (LTRs), especially young Gypsy‐type LTRs, are abundant in the peripheral CentO‐enriched regions and drive rice centromere expansion and evolution. Furthermore, high‐quality genome assembly and complete telomere‐to‐telomere (T2T) reference genome enable us to obtain more centromeric genome information despite mapping and cloning of centromere genes being challenging. We investigated the association between structural variations and gene expression in the rice centromere. A centromere gene, OsMAB, which positively regulates rice tiller number, was further confirmed by expression quantitative trait loci, haplotype analysis and clustered regularly interspaced palindromic repeats (CRISPR)/CRISPR‐associated protein 9 methods. By revealing the new insights into the evolutionary patterns and biological roles of rice centromeres, our finding will facilitate future research on centromere biology and crop improvement.