Rice (Oryza sativa L.) is a staple food for more than half of the world's population. To meet the ever-increasing demand for food, because of population growth and improved living standards, world rice production needs to double by 2030. The development of new elite rice varieties with high yield and superior quality is challenging for traditional breeding approaches, and new strategies need to be developed. Here, we report the successful development of new elite varieties by pyramiding major genes that significantly contribute to grain quality and yield from three parents over five years. The new varieties exhibit higher yield potential and better grain quality than their parental varieties and the China's leading super-hybrid rice, Liang-you-pai-jiu (LYP9 or Pei-ai 64S/93-11). Our results demonstrate that rational design is a powerful strategy for meeting the challenges of future crop breeding, particularly in pyramiding multiple complex traits.

The growing world population and shrinkage of arable land demand yield improvement of rice, one of the most important staple crops. To elucidate the genetic basis of yield and uncover its associated loci in rice, we resequenced the core recombinant inbred lines of Liang-YouPei-Jiu, the widely cultivated super hybrid rice, and constructed a high-resolution linkage map. We detected 43 yield-associated quantitative trait loci, of which 20 are unique. Based on the high-density physical map, the genome sequences of paternal variety 93-11 and maternal cultivar PA64s of Liang-You-Pei-Jiu were significantly improved. The large recombinant inbred line population combined with plentiful high-quality single nucleotide polymorphisms and insertions/deletions between parental genomes allowed us to fine-map two quantitative trait loci, qSN8 and qSPB1, and to identify days to heading8 and lax panicle1 as candidate genes, respectively. The quantitative trait locus qSN8 was further confirmed to be days to heading8 by a complementation test. Our study provided an ideal platform for molecular breeding by targeting and dissecting yieldassociated loci in rice.Oryza sativa | QTL dissection | genome sequence update R ice is one of the most important staple crops in the world and serves as a model for monocots (1). Currently, rice breeding faces the challenge of overcoming the yield plateau. All important agronomic traits would ultimately need to consider their impacts on the yield, which is linked to various growth and developmental components, such as tiller number, seed number and set, and grain weight, to name a few. A number of quantitative trait loci (QTLs) have been reported to control these components, including those revealed by map-based cloning studies, such as IPA1/WFP for tiller and spikelet numbers (2, 3); days to heading8 (DTH8)/Ghd8 and Ghd7 for heading date, plant height, and spikelet number (4, 5); Gn1 for spikelet number (6); GIF1 for seed set (7); and grain size3 (GS3) and GW5 for grain size and weight (8, 9). Although a series of QTLs for yield components have been cloned, elucidation of the genetic mechanisms underlying the inheritance of superior yield in super hybrid rice still has a long way to go.Hybrid rice has a notable contribution to yield improvement. Various commercialized hybrids are derived by crossing different varieties within or between two subspecies, Oryza sativa ssp. indica and ssp. japonica (10, 11). As a pioneer super hybrid rice, Liang-You-Pei-Jiu (LYP9) realized the target of 10.5 tons/ha in 2000 (12). LYP9 was developed by a cross of the paternal 93-11, an indica variety widely grown in China (13), and the maternal PA64s cultivar with a mixed genetic background of indica and javanica. To date, it has been widely cultivated for commercial production in China. Such a feature was thought to make LYP9 recombinant inbred lines (RILs) ideal materials for exploring molecular mechanisms underlying rice yield.Here, we constructed a high-density linkage map by resequencing the parents of LYP9 and 132 c...