Impact of genotype-calling methodologies on genome-wide association and genomic prediction in polyploids

Discovery and analysis of genetic variants underlying agriculturally important traits are key to molecular breeding of crops. Reduced representation approaches have provided cost-efficient genotyping using next-generation sequencing. However, accurate genotype calling from next-generation sequencing data is challenging, particularly in polyploid species due to their genome complexity. Recently developed Bayesian statistical methods implemented in available software packages, polyRAD, EBG, and updog, incorporate error rates and population parameters to accurately estimate allelic dosage across any ploidy. We used empirical and simulated data to evaluate the three Bayesian algorithms and demonstrated their impact on the power of genome-wide association study (GWAS) analysis and the accuracy of genomic prediction. We further incorporated uncertainty in allelic dosage estimation by testing continuous genotype calls and comparing their performance to discrete genotypes in GWAS and genomic prediction. We tested the genotype-calling methods using data from two autotetraploid species, Miscanthus sacchariflorus and Vaccinium corymbosum, and performed GWAS and genomic prediction. In the empirical study, the tested Bayesian genotype-calling algorithms differed in their downstream effects on GWAS and genomic prediction, with some showing advantages over others. Through subsequent simulation studies, we observed that at low read depth, polyRAD was advantageous in its effect on GWAS power and limit of false positives. Additionally, we found that continuous genotypes increased the accuracy of genomic prediction, by reducing genotyping error, particularly at low sequencing depth. Our results indicate that by using the Bayesian algorithm implemented in polyRAD and continuous genotypes, we can accurately and cost-efficiently implement GWAS and genomic prediction in polyploid crops.

Seed management using NGS technology to rapidly eliminate a deleterious allele from rice breeder seeds

Spontaneous mutations are stochastic phenomena that occur in every population. However, deleterious mutated allele present in seeds distributed to farmers must be detected and removed. Here, we eliminated undesirable mutations from the parent population in one generation through a strategy based on next-generation sequencing (NGS). This study dealt with a spontaneous albino mutant in the 'Hinohikari' rice variety grown at the Miyazaki Comprehensive Agricultural Experiment Station, Japan. The incidence of albinism in the population was 1.36%. NGS analysis revealed the genomic basis for differences between green and albino phenotypes. Every albino plant had a C insertion in the Snow-White Leaf1 (SWL1) gene on chromosome 4 causing a frameshift mutation. Selfing plants heterozygous for the mutant allele, swl1-R332P, resulted in a 3:1 green/albino ratio, confirming that a single recessive gene controls albinism. Ultrastructural leaf features in the swl1-R332P mutants displayed deformed chlorophyll-associated organelles in albino plants that were similar to those of previously described swl1 mutants. Detection of the causative gene and its confirmation using heterozygous progenies were completed within a year. The NGS technique outlined here facilitates rapid identification of spontaneous mutations that can occur in breeder seeds.

Training Population Optimization for Genomic Selection in Miscanthus

Miscanthus is a perennial grass with potential for lignocellulosic ethanol production. To ensure its utility for this purpose, breeding efforts should focus on increasing genetic diversity of the nothospecies Miscanthus × giganteus (M×g) beyond the single clone used in many programs. Germplasm from the corresponding parental species M. sinensis (Msi) and M. sacchariflorus (Msa) could theoretically be used as training sets for genomic prediction of M×g clones with optimal genomic estimated breeding values for biofuel traits. To this end, we first showed that subpopulation structure makes a substantial contribution to the genomic selection (GS) prediction accuracies within a 538-member diversity panel of predominately Msi individuals and a 598-member diversity panels of Msa individuals. We then assessed the ability of these two diversity panels to train GS models that predict breeding values in an interspecific diploid 216-member M×g F2 panel. Low and negative prediction accuracies were observed when various subsets of the two diversity panels were used to train these GS models. To overcome the drawback of having only one interspecific M×g F2 panel available, we also evaluated prediction accuracies for traits simulated in 50 simulated interspecific M×g F2 panels derived from different sets of Msi and diploid Msa parents. The results revealed that genetic architectures with common causal mutations across Msi and Msa yielded the highest prediction accuracies. Ultimately, these results suggest that the ideal training set should contain the same causal mutations segregating within interspecific M×g populations, and thus efforts should be undertaken to ensure that individuals in the training and validation sets are as closely related as possible.

Large-scale deployment of a rice 6 K SNP array for genetics and breeding applications

BACKGROUND

Fixed arrays of single nucleotide polymorphism (SNP) markers have advantages over reduced representation sequencing in their ease of data analysis, consistently higher call rates, and rapid turnaround times. A 6 K SNP array represents a cost-benefit "sweet spot" for routine genetics and breeding applications in rice. Selection of informative SNPs across species and subpopulations during chip design is essential to obtain useful polymorphism rates for target germplasm groups. This paper summarizes results from large-scale deployment of an Illumina 6 K SNP array for rice.

RESULTS

Design of the Illumina Infinium 6 K SNP chip for rice, referred to as the Cornell_6K_Array_Infinium_Rice (C6AIR), includes 4429 SNPs from re-sequencing data and 1571 SNP markers from previous BeadXpress 384-SNP sets, selected based on polymorphism rate and allele frequency within and between target germplasm groups. Of the 6000 attempted bead types, 5274 passed Illumina's production quality control. The C6AIR was widely deployed at the International Rice Research Institute (IRRI) for genetic diversity analysis, QTL mapping, and tracking introgressions and was intensively used at Cornell University for QTL analysis and developing libraries of interspecific chromosome segment substitution lines (CSSLs) between O. sativa and diverse accessions of O. rufipogon or O. meridionalis. Collectively, the array was used to genotype over 40,000 rice samples. A set of 4606 SNP markers was used to provide high quality data for O. sativa germplasm, while a slightly expanded set of 4940 SNPs was used for O. sativa X O. rufipogon populations. Biparental polymorphism rates were generally between 1900 and 2500 well-distributed SNP markers for indica x japonica or interspecific populations and between 1300 and 1500 markers for crosses within indica, while polymorphism rates were lower for pairwise crosses within U.S. tropical japonica germplasm. Recently, a second-generation array containing ~7000 SNP markers, referred to as the C7AIR, was designed by removing poor-performing SNPs from the C6AIR and adding markers selected to increase the utility of the array for elite tropical japonica material.

CONCLUSIONS

The C6AIR has been successfully used to generate rapid and high-quality genotype data for diverse genetics and breeding applications in rice, and provides the basis for an optimized design in the C7AIR.

Genetic variation of γ-tocopherol methyltransferase gene contributes to elevated α-tocopherol content in soybean seeds

BACKGROUND

Improvement of α-tocopherol content is an important breeding aim to increase the nutritional value of crops. Several efforts have been conducted to improve the α-tocopherol content in soybean [Glycine max (L.) Merr.] through transgenic technology by overexpressing genes related to α-tocopherol biosynthesis or through changes to crop management practices. Varieties with high α-tocopherol content have been identified in soybean germplasms. The heritability of this trait has been characterized in a cross between high α-tocopherol variety Keszthelyi Aproszemu Sarga (KAS) and low α-tocopherol variety Ichihime. In this study, the genetic mechanism of the high α-tocopherol content trait of KAS was elucidated.

RESULTS

Through QTL analysis and fine mapping in populations from a cross between KAS and a Japanese variety Ichihime, we identified γ-TMT3, which encodes γ-tocopherol methyltransferase, as a candidate gene responsible for high α-tocopherol concentration in KAS. Several nucleotide polymorphisms including two nonsynonymous mutations were found in the coding region of γ-TMT3 between Ichihime and KAS, but none of which was responsible for the difference in α-tocopherol concentration. Therefore, we focused on transcriptional regulation of γ-TMT3 in developing seeds and leaves. An F5 line that was heterozygous for the region containing γ-TMT3 was self-pollinated. From among the progeny, plants that were homozygous at the γ-TMT3 locus were chosen for further evaluation. The expression level of γ-TMT3 was higher both in developing seeds and leaves of plants homozygous for the γ-TMT3 allele from KAS. The higher expression level was closely correlated with high α-tocopherol content in developing seeds. We generated transgenic Arabidopsis plants harboring GUS gene under the control of γ-TMT3 promoter from KAS or Ichihime. The GUS activity assay showed that the activity of γ-TMT3 promoter from KAS was higher than that of Ichihime.

CONCLUSIONS

The genetic variation in γ-TMT3, which plays a major role in determining α-tocopherol concentration, provides significant information about the regulation of tocopherol biosynthesis in soybean seeds. This knowledge will help breeding programs to develop new soybean varieties with high α-tocopherol content.