Two populations of interspecific introgression lines (ILs) in a common recurrent parent were developed for use in pre-breeding and QTL mapping. The ILs were derived from crosses between cv Curinga, a tropical japonica upland cultivar, and two different wild donors, Oryza meridionalis Ng. accession (W2112) and Oryza rufipogon Griff. accession (IRGC 105491). The lines were genotyped using genotyping-by-sequencing (GBS) and SSRs. The 32 Curinga/O. meridionalis ILs contain 76.73 % of the donor genome in individual introgressed segments, and each line has an average of 94.9 % recurrent parent genome. The 48 Curinga/O. rufipogon ILs collectively contain 97.6 % of the donor genome with an average of 89.9 % recurrent parent genome per line. To confirm that these populations were segregating for traits of interest, they were phenotyped for pericarp color in the greenhouse and for four agronomic traits—days to flowering, plant height, number of tillers, and number of panicles—in an upland field environment. Seeds from these IL libraries and the accompanying GBS datasets are publicly available and represent valuable genetic resources for exploring the genetics and breeding potential of rice wild relatives.Electronic supplementary materialThe online version of this article (doi:10.1007/s11032-015-0276-7) contains supplementary material, which is available to authorized users.

Number of spikelets per panicle (NSP) is a key trait to increase yield potential in rice (O. sativa). The architecture of the rice inflorescence which is mainly determined by the length and number of primary (PBL and PBN) and secondary (SBL and SBN) branches can influence NSP. Although several genes controlling panicle architecture and NSP in rice have been identified, there is little evidence of (i) the genetic control of panicle architecture and NSP in different environments and (ii) the presence of stable genetic associations with panicle architecture across environments. This study combines image phenotyping of 225 accessions belonging to a genetic diversity array of indica rice grown under irrigated field condition in two different environments and Genome Wide Association Studies (GWAS) based on the genotyping of the diversity panel, providing 83,374 SNPs. Accessions sown under direct seeding in one environement had reduced Panicle Length (PL), NSP, PBN, PBL, SBN, and SBL compared to those established under transplanting in the second environment. Across environments, NSP was significantly and positively correlated with PBN, SBN and PBL. However, the length of branches (PBL and SBL) was not significantly correlated with variables related to number of branches (PBN and SBN), suggesting independent genetic control. Twenty- three GWAS sites were detected with P ≤ 1.0E-04 and 27 GWAS sites with p ≤ 5.9E−04. We found 17 GWAS sites related to NSP, 10 for PBN and 11 for SBN, 7 for PBL and 11 for SBL. This study revealed new regions related to NSP, but only three associations were related to both branching number (PBN and SBN) and NSP. Two GWAS sites associated with SBL and SBN were stable across contrasting environments and were not related to genes previously reported. The new regions reported in this study can help improving NSP in rice for both direct seeded and transplanted conditions. The integrated approach of high-throughput phenotyping, multi-environment field trials and GWAS has the potential to dissect complex traits, such as NSP, into less complex traits and to match single nucleotide polymorphisms with relevant function under different environments, offering a potential use for molecular breeding.

Genomic selection (GS) is a promising strategy for enhancing genetic gain. We investigated the accuracy of genomic estimated breeding values (GEBV) in four inter-related synthetic populations that underwent several cycles of recurrent selection in an upland rice-breeding program. A total of 343 S2:4 lines extracted from those populations were phenotyped for flowering time, plant height, grain yield and panicle weight, and genotyped with an average density of one marker per 44.8 kb. The relative effect of the linkage disequilibrium (LD) and minor allele frequency (MAF) thresholds for selecting markers, the relative size of the training population (TP) and of the validation population (VP), the selected trait and the genomic prediction models (frequentist and Bayesian) on the accuracy of GEBVs was investigated in 540 cross validation experiments with 100 replicates. The effect of kinship between the training and validation populations was tested in an additional set of 840 cross validation experiments with a single genomic prediction model. LD was high (average r2 = 0.59 at 25 kb) and decreased slowly, distribution of allele frequencies at individual loci was markedly skewed toward unbalanced frequencies (MAF average value 15.2% and median 9.6%), and differentiation between the four synthetic populations was low (FST ≤0.06). The accuracy of GEBV across all cross validation experiments ranged from 0.12 to 0.54 with an average of 0.30. Significant differences in accuracy were observed among the different levels of each factor investigated. Phenotypic traits had the biggest effect, and the size of the incidence matrix had the smallest. Significant first degree interaction was observed for GEBV accuracy between traits and all the other factors studied, and between prediction models and LD, MAF and composition of the TP. The potential of GS to accelerate genetic gain and breeding options to increase the accuracy of predictions are discussed.