Tuong-Vi Cao

Accuracy of Genomic Selection in a Rice Synthetic Population Developed for Recurrent Selection 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.

A Unified and Comprehensible View of Parametric and Kernel Methods for Genomic Prediction with Application to Rice

One objective of this study was to provide readers with a clear and unified understanding of parametric statistical and kernel methods, used for genomic prediction, and to compare some of these in the context of rice breeding for quantitative traits. Furthermore, another objective was to provide a simple and user-friendly R package, named KRMM, which allows users to perform RKHS regression with several kernels. After introducing the concept of regularized empirical risk minimization, the connections between well-known parametric and kernel methods such as Ridge regression [i.e., genomic best linear unbiased predictor (GBLUP)] and reproducing kernel Hilbert space (RKHS) regression were reviewed. Ridge regression was then reformulated so as to show and emphasize the advantage of the kernel “trick” concept, exploited by kernel methods in the context of epistatic genetic architectures, over parametric frameworks used by conventional methods. Some parametric and kernel methods; least absolute shrinkage and selection operator (LASSO), GBLUP, support vector machine regression (SVR) and RKHS regression were thereupon compared for their genomic predictive ability in the context of rice breeding using three real data sets. Among the compared methods, RKHS regression and SVR were often the most accurate methods for prediction followed by GBLUP and LASSO. An R function which allows users to perform RR-BLUP of marker effects, GBLUP and RKHS regression, with a Gaussian, Laplacian, polynomial or ANOVA kernel, in a reasonable computation time has been developed. Moreover, a modified version of this function, which allows users to tune kernels for RKHS regression, has also been developed and parallelized for HPC Linux clusters. The corresponding KRMM package and all scripts have been made publicly available.

Breeding for outcrossing ability in rice, to enhance seed production for hybrid rice cropping

Background: Adoption of the hybrid rice varieties by farmers is often impaired by the high price of hybrid seed, due to low yields in hybrid seed production fields. Female outcrossing ability (FOA) and female hybrid seed production ability (FHSPA), defined as the rate of filled spikelets of the male sterile (MS) line and as its grain yield under outcrossing, respectively, determine plant traits for hybrid seed yield. Breeding for FOA and FHSPA in rice has suffered from the lack of a high throughput phenotyping method and the inbred breeding approach used for the development of MS lines. We developed an innovative hybrid rice breeding strategy that uses the monogenic recessive male-sterility gene ms-IR36 for the reciprocal recurrent improvement of maintainer and restorer populations. Results: High throughput screening for FOA and FHSPA can be achieved by scoring the grain weight of MS plants and the grain yield of fertile plants of progenies extracted from breeding populations segregating for the ms-IR36 gene. Using this phenotyping method in seven field trials, each involving several hundred entries, we revealed a very broad diversity for FOA (ranging from zero to 89%) and FHSPA, within the F3 progenies of bi-parental crosses and within S1 and S2 progenies extracted from different breeding populations. The seven experiments produced convergent results and heritabilities of 0.59-0.90 for FHSPA and 0.45-0.72 for FOA. Correlations between FHSPA and FOA were tight and highly significant. Correlations were looser between FHSPA and grain yield of the selfed fertile sibling (GW-MF). Correlations between FOA and GW-MF were not significant. Tight significant correlation was also observed between FHSPA of S1 lines and S2 lines extracted from the former. Conclusion: Population breeding through recurrent selection, using the ms-IR36 gene as a tool for both recombination and seed production for testcrossing, is a favorable framework for harnessing rice genetic diversity for FHSPA. Rapid and cost-effective genetic gain for hybrid seed production can be achieved using results of the ms-IR36 gene mediated test cross seed production process as FHSPA early screening.

Correction: Accuracy of Genomic Selection in a Rice Synthetic Population Developed for Recurrent Selection Breeding.

[This corrects the article DOI: 10.1371/journal.pone.0136594.].

DHOEM: a statistical simulation software for simulating new markers in real SNP marker data.

BackgroundNumerous simulation tools based on specific assumptions have been proposed to simulate populations. Here we present a simulation tool named DHOEM (densification of haplotypes by loess regression and maximum likelihood) which is free from population assumptions and simulates new markers in real SNP marker data. The main objective of DHOEM is to generate a new population, which incorporates real and simulated SNP by statistical learning from an initial population, which match the realized features of the latter.ResultsTo demonstrate DHOEM’s abilities, we used a sample of 704 haplotypes for 12 chromosomes with 8336 SNP from a synthetic population, used for breeding upland rice in Latin America. The distributions of allele frequencies, pairwise SNP LD coefficients and data structures, before and after marker densification of the associated marker data set, were shown to be in relatively good agreement at moderate degrees of marker densification. DHOEM is a user-friendly tool that allows the user to specify the level of marker density desired, with a user defined minor allele frequency (MAF) limit, which is produced in a reasonable computation time.ConclusionsDHOEM is a user-friendly and useful tool for simulation and methodological studies in quantitative genetics and breeding.