Future demands for increased productivity and resilience to abiotic/biotic stresses of major crops require new technologies of breeding by design (BBD) built on massive information from functional and population genomics research. A novel strategy of breeding by selective introgression (BBSI) has been proposed and practiced for simultaneous improvement, genetic dissection and allele mining of complex traits to realize BBD. BBSI has three phases: a) developing large numbers of trait-specific introgression lines (ILs) using backcross breeding in elite genetic backgrounds as the material platform of BBD; b) efficiently identifying genes or quantitative trait loci (QTL) and mining desirable alleles affecting different target traits from diverse donors as the information platform of BBD; and c) developing superior cultivars by BBD using designed QTL pyramiding or marker-assisted recurrent selection. Phase (a) has been implemented massively in rice by many Chinese research institutions and IRRI, resulting in the development of many new green super rice cultivars plus large numbers of ILs in 30 + elite genetic backgrounds. Phase (b) has been demonstrated in a series of proof-of-concept studies of high-efficiency genetic dissection of rice yield and tolerance to abiotic stresses using ILs and DNA markers. Phase (c) has also been implemented by designed QTL pyramiding, resulting in a prototype of BBD in several successful cases. The BBSI strategy can be easily extended for simultaneous trait improvement, efficient gene and QTL discovery and allele mining of complex traits using advanced breeding lines from crosses between a common “backbone” parent and a set of elite parents in conventional pedigree breeding programs. BBSI can be relatively easily adopted by breeding programs with small budgets, but the BBSI-based BBD strategy can be fully and more efficiently implemented by large seed companies with sufficient capacity.

The development of green super rice varieties with improved nutrient use efficiency (NuUE) is a vital target area to increase yield and make it more stable under rainfed conditions. In the present study, we followed an early backcross (BC) breeding approach by using a high-yielding and widely adapted Xian variety, Weed Tolerant Rice 1 (WTR-1), as a recipient and a Geng variety, Hao-An-Nong (HAN), as a donor. Starting from the BC₁F₂ generation, the BC population went through one generation of selection under irrigated, low-input, and rainfed conditions, followed by four consecutive generations of screening and selection for high grain yield (GY) under six different nutrient conditions (NPK, 75N, -N, -P, -NP, and -NPK), leading to the development of 230 BC₁F₆ introgression lines (ILs). These 230 ILs were evaluated under the same six nutrient conditions for 13 agro-morphological and grain yield component traits in comparison to four checks and parents. Significant trait variations were observed between the treatments and ILs. Positive correlations were identified for GY with biomass, panicle length, flag-leaf area, flag-leaf width, filled grain number per panicle, 1000-grain weight, and tiller number under -N, -P, -NP, and -NPK conditions. Out of 230 ILs, 12 were identified as promising under two or more nutrient deficiency conditions. The results demonstrated an efficient inter-subspecific BC breeding procedure with a first round of selection under rainfed-drought conditions, followed by four generations of progeny testing for yield performance under six nutrient conditions. The promising ILs can be useful resources for molecular genetic dissection and understanding the physiological mechanisms of NuUE.

The backcross (BC) breeding strategy has been increasingly used for developing high yielding varieties with improved abiotic stress tolerances in rice. In this study, 189 Huang-Hua-Zhan (HHZ) introgression lines (ILs) developed from three different selection schemes were evaluated for yield related traits under drought stress and non-stress conditions in the target and off-season winter nursery environments to assess the selection efficiency of BC breeding for improving different complex traits, and led us to five important results. The first result indicated that the primary target traits should be selected first in the target environments (TEs) in order to achieve the maximum genetic gain. Secondly, BC breeding for drought tolerance (DT) in rice was almost equally effective by strong phenotypic selection in the main target environments and in the winter-season of Hainan. Thirdly, exploiting genetic diversity in the subspecific gene pools is of great importance for future genetic improvement of complex traits in rice. Fourthly, considerable genetic gain can be effectively achieved by selection for secondary target traits among the ILs with the primary traits. Finally, the developed ILs provide useful materials for future genetic/genomic dissection and molecular breeding of complex traits.