From Gene-Based Breeding to Haplotype-Based Breeding for Rice (Oryza sativa) Improvement in the Genomics Era

Rice (Oryza sativa L.) improvement increasingly uses genome-scale variation, but breeding decisions remain trait-oriented and depend on genetic architecture, recipient background, target environment, and selection resources. This review assesses when phased haplotype information adds value beyond causal-gene markers, single-nucleotide polymorphism assays, phenotypic selection, and genomic prediction. Reported configurations for yield, grain quality, nutrition, biotic resistance, and abiotic-stress adaptation are evaluated according to population structure, carrier frequency, effect uncertainty, haplotype-by-background interaction, haplotype-by-environment interaction, linkage drag, transferability, assay availability, and current breeding use. A three-tier hierarchy classifies configurations as Candidate, Validated, or Breeder-Ready. Most reported favorable haplotypes remain Candidate targets, fewer have controlled genetic validation, and few sequence-defined multi-variant haplotypes have documented use in active breeding programs. Practical deployment requires stable sequence nomenclature, accessible donors, diagnostic assays, elite-background and multi-environment evaluation, cost analysis, and donor-segment recovery. Graph pangenomes, phased long-read assemblies, causal reconstruction, and prospective comparisons with current selection methods can improve future assessment. Haplotype information has greatest value when phase improves a defined breeding decision and increases target-environment performance, annual genetic gain, selection efficiency, or cultivar delivery.