Transcriptome-based prediction for polygenic traits in rice using different gene subsets.

BACKGROUND: Transcriptome-based prediction of complex phenotypes is a relatively new statistical method that links genetic variation to phenotypic variation. The selection of large-effect genes based on a priori biological knowledge is beneficial for predicting oligogenic traits; however, such a simple gene selection method is not applicable to polygenic traits because causal genes or large-effect loci are often unknown. Here, we used several gene-level features and tested whether it was possible to select a gene subset that resulted in better predictive ability than using all genes for predicting a polygenic trait. RESULTS: Using the phenotypic values of shoot and root traits and transcript abundances in leaves and roots of 57 rice accessions, we evaluated the predictive abilities of the transcriptome-based prediction models. Leaf transcripts predicted shoot phenotypes, such as plant height, more accurately than root transcripts, whereas root transcripts predicted root phenotypes, such as crown root length, more accurately than leaf transcripts. Furthermore, we used the following three features to train the prediction model: (1) tissue specificity of the transcripts, (2) ontology annotations, and (3) co-expression modules for selecting gene subsets. Although models trained by a gene subset often resulted in lower predictive abilities than the model trained by all genes, some gene subsets showed improved predictive ability. For example, using genes expressed in roots but not in leaves, the predictive ability for crown root diameter was improved by more than 10% (R2 = 0.59 when using all genes; R2 = 0.66, using 1,554 root-specifically expressed genes). Similarly, genes annotated as "gibberellic acid sensitivity" showed higher predictive ability than using all genes for root dry weight. CONCLUSIONS: Our results highlight both the possibility and difficulty of selecting an appropriate gene subset to predict polygenic traits from transcript abundance, given the current biological knowledge and information. Further integration of multiple sources of information, as well as improvements in gene characterization, may enable the selection of an optimal gene set for the prediction of polygenic phenotypes.

Genome- and Transcriptome-wide Association Studies to Discover Candidate Genes for Diverse Root Phenotypes in Cultivated Rice.

Root system architecture plays a crucial role in nutrient and water absorption during rice production. Genetic improvement of the rice root system requires elucidating its genetic control. Genome-wide association studies (GWASs) have identified genomic regions responsible for rice root phenotypes. However, candidate gene prioritization around the peak region often suffers from low statistical power and resolution. Transcriptomics enables other statistical mappings, such as transcriptome-wide association study (TWAS) and expression GWAS (eGWAS), which improve candidate gene identification by leveraging the natural variation of the expression profiles. To explore the genes responsible for root phenotypes, we conducted GWAS, TWAS, and eGWAS for 12 root phenotypes in 57 rice accessions using 427,751 single nucleotide polymorphisms (SNPs) and the expression profiles of 16,901 genes expressed in the roots. The GWAS identified three significant peaks, of which the most significant peak responsible for seven root phenotypes (crown root length, crown root surface area, number of crown root tips, lateral root length, lateral root surface area, lateral root volume, and number of lateral root tips) was detected at 6,199,732 bp on chromosome 8. In the most significant GWAS peak region, OsENT1 was prioritized as the most plausible candidate gene because its expression profile was strongly negatively correlated with the seven root phenotypes. In addition to OsENT1, OsEXPA31, OsSPL14, OsDEP1, and OsDEC1 were identified as candidate genes responsible for root phenotypes using TWAS. Furthermore, a cis-eGWAS peak SNP was detected for OsDjA6, which showed the eighth strongest association with lateral root volume in the TWAS. The cis-eGWAS peak SNP for OsDjA6 was in strong linkage disequilibrium (LD) with a GWAS peak SNP on the same chromosome for lateral root volume and in perfect LD with another SNP variant in a putative cis-element at the 518 bp upstream of the gene. These candidate genes provide new insights into the molecular breeding of root system architecture.

Genomic prediction of tocochromanols in exotic-derived maize.

Tocochromanols (vitamin E) are an essential part of the human diet. Plant products, including maize (Zea mays L.) grain, are the major dietary source of tocochromanols; therefore, breeding maize with higher vitamin content (biofortification) could improve human nutrition. Incorporating exotic germplasm in maize breeding for trait improvement including biofortification is a promising approach and an important research topic. However, information about genomic prediction of exotic-derived lines using available training data from adapted germplasm is limited. In this study, genomic prediction was systematically investigated for nine tocochromanol traits within both an adapted (Ames Diversity Panel [AP]) and an exotic-derived (Backcrossed Germplasm Enhancement of Maize [BGEM]) maize population. Although prediction accuracies up to 0.79 were achieved using genomic best linear unbiased prediction (gBLUP) when predicting within each population, genomic prediction of BGEM based on an AP training set resulted in low prediction accuracies. Optimal training population (OTP) design methods fast and unique representative subset selection (FURS), maximization of connectedness and diversity (MaxCD), and partitioning around medoids (PAM) were adapted for inbreds and, along with the methods mean coefficient of determination (CDmean) and mean prediction error variance (PEVmean), often improved prediction accuracies compared with random training sets of the same size. When applied to the combined population, OTP designs enabled successful prediction of the rest of the exotic-derived population. Our findings highlight the importance of leveraging genotype data in training set design to efficiently incorporate new exotic germplasm into a plant breeding program.

Leveraging prior biological knowledge improves prediction of tocochromanols in maize grain.

With an essential role in human health, tocochromanols are mostly obtained by consuming seed oils; however, the vitamin E content of the most abundant tocochromanols in maize (Zea mays L.) grain is low. Several large-effect genes with cis-acting variants affecting messenger RNA (mRNA) expression are mostly responsible for tocochromanol variation in maize grain, with other relevant associated quantitative trait loci (QTL) yet to be fully resolved. Leveraging existing genomic and transcriptomic information for maize inbreds could improve prediction when selecting for higher vitamin E content. Here, we first evaluated a multikernel genomic best linear unbiased prediction (MK-GBLUP) approach for modeling known QTL in the prediction of nine tocochromanol grain phenotypes (12-21 QTL per trait) within and between two panels of 1,462 and 242 maize inbred lines. On average, MK-GBLUP models improved predictive abilities by 7.0-13.6% when compared with GBLUP. In a second approach with a subset of 545 lines from the larger panel, the highest average improvement in predictive ability relative to GBLUP was achieved with a multi-trait GBLUP model (15.4%) that had a tocochromanol phenotype and transcript abundances in developing grain for a few large-effect candidate causal genes (1-3 genes per trait) as multiple response variables. Taken together, our study illustrates the enhancement of prediction models when informed by existing biological knowledge pertaining to QTL and candidate causal genes.

Combining GWAS and TWAS to identify candidate causal genes for tocochromanol levels in maize grain.

Tocochromanols (tocopherols and tocotrienols, collectively vitamin E) are lipid-soluble antioxidants important for both plant fitness and human health. The main dietary sources of vitamin E are seed oils that often accumulate high levels of tocopherol isoforms with lower vitamin E activity. The tocochromanol biosynthetic pathway is conserved across plant species but an integrated view of the genes and mechanisms underlying natural variation of tocochromanol levels in seed of most cereal crops remains limited. To address this issue, we utilized the high mapping resolution of the maize Ames panel of ∼1,500 inbred lines scored with 12.2 million single-nucleotide polymorphisms to generate metabolomic (mature grain tocochromanols) and transcriptomic (developing grain) data sets for genetic mapping. By combining results from genome- and transcriptome-wide association studies, we identified a total of 13 candidate causal gene loci, including 5 that had not been previously associated with maize grain tocochromanols: 4 biosynthetic genes (arodeH2 paralog, dxs1, vte5, and vte7) and a plastid S-adenosyl methionine transporter (samt1). Expression quantitative trait locus (eQTL) mapping of these 13 gene loci revealed that they are predominantly regulated by cis-eQTL. Through a joint statistical analysis, we implicated cis-acting variants as responsible for colocalized eQTL and GWAS association signals. Our multiomics approach provided increased statistical power and mapping resolution to enable a detailed characterization of the genetic and regulatory architecture underlying tocochromanol accumulation in maize grain and provided insights for ongoing biofortification efforts to breed and/or engineer vitamin E and antioxidant levels in maize and other cereals.

Genomic prediction of zinc-biofortification potential in rice gene bank accessions.

KEY MESSAGE: A genomic prediction model successfully predicted grain Zn concentrations in 3000 gene bank accessions and this was verified experimentally with selected potential donors having high on-farm grain-Zn in Madagascar. Increasing zinc (Zn) concentrations in edible parts of food crops, an approach termed Zn-biofortification, is a global breeding objective to alleviate micro-nutrient malnutrition. In particular, infants in countries like Madagascar are at risk of Zn deficiency because their dominant food source, rice, contains insufficient Zn. Biofortified rice varieties with increased grain Zn concentrations would offer a solution and our objective is to explore the genotypic variation present among rice gene bank accessions and to possibly identify underlying genetic factors through genomic prediction and genome-wide association studies (GWAS). A training set of 253 rice accessions was grown at two field sites in Madagascar to determine grain Zn concentrations and grain yield. A multi-locus GWAS analysis identified eight loci. Among these, QTN_11.3 had the largest effect and a rare allele increased grain Zn concentrations by 15%. A genomic prediction model was developed from the above training set to predict Zn concentrations of 3000 sequenced rice accessions. Predicted concentrations ranged from 17.1 to 40.2 ppm with a prediction accuracy of 0.51. An independent confirmation with 61 gene bank seed samples provided high correlations (r = 0.74) between measured and predicted values. Accessions from the aus sub-species had the highest predicted grain Zn concentrations and these were confirmed in additional field experiments, with one potential donor having more than twice the grain Zn compared to a local check variety. We conclude utilizing donors from the aus sub-species and employing genomic selection during the breeding process is the most promising approach to raise grain Zn concentrations in rice.

Phenotyping of a rice (Oryza sativa L.) association panel identifies loci associated with tolerance to low soil fertility on smallholder farm conditions in Madagascar.

Rice (Oryza sativa L.) is a staple food of Madagascar, where per capita rice consumption is among the highest worldwide. Rice in Madagascar is mainly grown on smallholder farms on soils with low fertility and in the absence of external inputs such as mineral fertilizers. Consequently, rice productivity remains low and the gap between rice production and consumption is widening at the national level. This study evaluates genetic resources imported from the IRRI rice gene bank to identify potential donors and loci associated with low soil fertility tolerance (LFT) that could be utilized in improving rice yield under local cultivation conditions. Accessions were grown on-farm without fertilizer inputs in the central highlands of Madagascar. A Genome-wide association study (GWAS) identified quantitative trait loci (QTL) for total panicle weight per plant, straw weight, total plant biomass, heading date and plant height. We detected loci at locations of known major genes for heading date (hd1) and plant height (sd1), confirming the validity of GWAS procedures. Two QTLs for total panicle weight were detected on chromosomes 5 (qLFT5) and 11 (qLFT11) and superior panicle weight was conferred by minor alleles. Further phenotyping under P and N deficiency suggested qLFT11 to be related to preferential resource allocation to root growth under nutrient deficiency. A donor (IRIS 313-11949) carrying both minor advantageous alleles was identified and crossed to a local variety (X265) lacking these alleles to initiate variety development through a combination of marker-assisted selection with selection on-farm in the target environment rather than on-station as typically practiced.

Transcriptome-wide association and prediction for carotenoids and tocochromanols in fresh sweet corn kernels.

Sweet corn (Zea mays L.) is consistently one of the most highly consumed vegetables in the United States, providing a valuable opportunity to increase nutrient intake through biofortification. Significant variation for carotenoid (provitamin A, lutein, zeaxanthin) and tocochromanol (vitamin E, antioxidants) levels is present in temperate sweet corn germplasm, yet previous genome-wide association studies (GWAS) of these traits have been limited by low statistical power and mapping resolution. Here, we employed a high-quality transcriptomic dataset collected from fresh sweet corn kernels to conduct transcriptome-wide association studies (TWAS) and transcriptome prediction studies for 39 carotenoid and tocochromanol traits. In agreement with previous GWAS findings, TWAS detected significant associations for four causal genes, ß-carotene hydroxylase (crtRB1), lycopene epsilon cyclase (lcyE), γ-tocopherol methyltransferase (vte4), and homogentisate geranylgeranyltransferase (hggt1) on a transcriptome-wide level. Pathway-level analysis revealed additional associations for deoxy-xylulose synthase2 (dxs2), diphosphocytidyl methyl erythritol synthase2 (dmes2), cytidine methyl kinase1 (cmk1), and geranylgeranyl hydrogenase1 (ggh1), of which, dmes2, cmk1, and ggh1 have not previously been identified through maize association studies. Evaluation of prediction models incorporating genome-wide markers and transcriptome-wide abundances revealed a trait-dependent benefit to the inclusion of both genomic and transcriptomic data over solely genomic data, but both transcriptome- and genome-wide datasets outperformed a priori candidate gene-targeted prediction models for most traits. Altogether, this study represents an important step toward understanding the role of regulatory variation in the accumulation of vitamins in fresh sweet corn kernels.

From gene banks to farmer's fields: using genomic selection to identify donors for a breeding program in rice to close the yield gap on smallholder farms.

KEY MESSAGE: Despite phenotyping the training set under unfavorable conditions on smallholder farms in Madagascar, we were able to successfully apply genomic prediction to select donors among gene bank accessions. Poor soil fertility and low fertilizer application rates are main reasons for the large yield gap observed for rice produced in sub-Saharan Africa. Traditional varieties that are preserved in gene banks were shown to possess traits and alleles that would improve the performance of modern variety under such low-input conditions. How to accelerate the utilization of gene bank resources in crop improvement is an unresolved question and here our objective was to test whether genomic prediction could aid in the selection of promising donors. A subset of the 3,024 sequenced accessions from the IRRI rice gene bank was phenotyped for yield and agronomic traits for two years in unfertilized farmers' fields in Madagascar, and based on these data, a genomic prediction model was developed. This model was applied to predict the performance of the entire set of 3024 accessions, and the top predicted performers were sent to Madagascar for confirmatory trials. The prediction accuracies ranged from 0.10 to 0.30 for grain yield, from 0.25 to 0.63 for straw biomass, to 0.71 for heading date. Two accessions have subsequently been utilized as donors in rice breeding programs in Madagascar. Despite having conducted phenotypic evaluations under challenging conditions on smallholder farms, our results are encouraging as the prediction accuracy realized in on-farm experiments was in the range of accuracies achieved in on-station studies. Thus, we could provide clear empirical evidence on the value of genomic selection in identifying suitable genetic resources for crop improvement, if genotypic data are available.

High-resolution genome-wide association study pinpoints metal transporter and chelator genes involved in the genetic control of element levels in maize grain.

Despite its importance to plant function and human health, the genetics underpinning element levels in maize grain remain largely unknown. Through a genome-wide association study in the maize Ames panel of nearly 2,000 inbred lines that was imputed with ∼7.7 million SNP markers, we investigated the genetic basis of natural variation for the concentration of 11 elements in grain. Novel associations were detected for the metal transporter genes rte2 (rotten ear2) and irt1 (iron-regulated transporter1) with boron and nickel, respectively. We also further resolved loci that were previously found to be associated with one or more of five elements (copper, iron, manganese, molybdenum, and/or zinc), with two metal chelator and five metal transporter candidate causal genes identified. The nas5 (nicotianamine synthase5) gene involved in the synthesis of nicotianamine, a metal chelator, was found associated with both zinc and iron and suggests a common genetic basis controlling the accumulation of these two metals in the grain. Furthermore, moderate predictive abilities were obtained for the 11 elemental grain phenotypes with two whole-genome prediction models: Bayesian Ridge Regression (0.33-0.51) and BayesB (0.33-0.53). Of the two models, BayesB, with its greater emphasis on large-effect loci, showed ∼4-10% higher predictive abilities for nickel, molybdenum, and copper. Altogether, our findings contribute to an improved genotype-phenotype map for grain element accumulation in maize.

SpaTemHTP: A Data Analysis Pipeline for Efficient Processing and Utilization of Temporal High-Throughput Phenotyping Data.

The rapid development of phenotyping technologies over the last years gave the opportunity to study plant development over time. The treatment of the massive amount of data collected by high-throughput phenotyping (HTP) platforms is however an important challenge for the plant science community. An important issue is to accurately estimate, over time, the genotypic component of plant phenotype. In outdoor and field-based HTP platforms, phenotype measurements can be substantially affected by data-generation inaccuracies or failures, leading to erroneous or missing data. To solve that problem, we developed an analytical pipeline composed of three modules: detection of outliers, imputation of missing values, and mixed-model genotype adjusted means computation with spatial adjustment. The pipeline was tested on three different traits (3D leaf area, projected leaf area, and plant height), in two crops (chickpea, sorghum), measured during two seasons. Using real-data analyses and simulations, we showed that the sequential application of the three pipeline steps was particularly useful to estimate smooth genotype growth curves from raw data containing a large amount of noise, a situation that is potentially frequent in data generated on outdoor HTP platforms. The procedure we propose can handle up to 50% of missing values. It is also robust to data contamination rates between 20 and 30% of the data. The pipeline was further extended to model the genotype time series data. A change-point analysis allowed the determination of growth phases and the optimal timing where genotypic differences were the largest. The estimated genotypic values were used to cluster the genotypes during the optimal growth phase. Through a two-way analysis of variance (ANOVA), clusters were found to be consistently defined throughout the growth duration. Therefore, we could show, on a wide range of scenarios, that the pipeline facilitated efficient extraction of useful information from outdoor HTP platform data. High-quality plant growth time series data is also provided to support breeding decisions. The R code of the pipeline is available at https://github.com/ICRISAT-GEMS/SpaTemHTP.

Automated discretization of 'transpiration restriction to increasing VPD' features from outdoors high-throughput phenotyping data.

BACKGROUND: Restricting transpiration under high vapor pressure deficit (VPD) is a promising water-saving trait for drought adaptation. However, it is often measured under controlled conditions and at very low throughput, unsuitable for breeding. A few high-throughput phenotyping (HTP) studies exist, and have considered only maximum transpiration rate in analyzing genotypic differences in this trait. Further, no study has precisely identified the VPD breakpoints where genotypes restrict transpiration under natural conditions. Therefore, outdoors HTP data (15 min frequency) of a chickpea population were used to automate the generation of smooth transpiration profiles, extract informative features of the transpiration response to VPD for optimal genotypic discretization, identify VPD breakpoints, and compare genotypes. RESULTS: Fifteen biologically relevant features were extracted from the transpiration rate profiles derived from load cells data. Genotypes were clustered (C1, C2, C3) and 6 most important features (with heritability > 0.5) were selected using unsupervised Random Forest. All the wild relatives were found in C1, while C2 and C3 mostly comprised high TE and low TE lines, respectively. Assessment of the distinct p-value groups within each selected feature revealed highest genotypic variation for the feature representing transpiration response to high VPD condition. Sensitivity analysis on a multi-output neural network model (with R of 0.931, 0.944, 0.953 for C1, C2, C3, respectively) found C1 with the highest water saving ability, that restricted transpiration at relatively low VPD levels, 56% (i.e. 3.52 kPa) or 62% (i.e. 3.90 kPa), depending whether the influence of other environmental variables was minimum or maximum. Also, VPD appeared to have the most striking influence on the transpiration response independently of other environment variable, whereas light, temperature, and relative humidity alone had little/no effect. CONCLUSION: Through this study, we present a novel approach to identifying genotypes with drought-tolerance potential, which overcomes the challenges in HTP of the water-saving trait. The six selected features served as proxy phenotypes for reliable genotypic discretization. The wild chickpeas were found to limit water-loss faster than the water-profligate cultivated ones. Such an analytic approach can be directly used for prescriptive breeding applications, applied to other traits, and help expedite maximized information extraction from HTP data.

Effect of salt tolerance on biomass production in a large population of sorghum accessions.

Salinity causes major reductions in cultivated land area, crop productivity, and crop quality, and salt-tolerant crops have been required to sustain agriculture in salinized areas. The annual C4 crop plant Sorghum bicolor (L.) Moench is salt tolerant, with large variation among accessions. Sorghum's salt tolerance is often evaluated during early growth, but such evaluations are weakly related to overall performance. Here, we evaluated salt tolerance of 415 sorghum accessions grown in saline soil (0, 50, 100, and 150 mM NaCl) for 3 months. Some accessions produced up to 400 g per plant of biomass and showed no growth inhibition at 50 mM NaCl. Our analysis indicated that the genetic factors that affected biomass production under 100 mM salt stress were more different from those without salt stress, comparing to the differences between those under 50 mM and 100 mM salt stress. A genome-wide association study for salt tolerance identified two single-nucleotide polymorphisms (SNPs) that were significantly associated with biomass production, only at 50 mM NaCl. Additionally, two SNPs were significantly associated with salt tolerance index as an indicator for growth response of each accession to salt stress. Our results offer candidate genetic resources and SNP markers for breeding salt-tolerant sorghum.

Bayesian optimization for genomic selection: a method for discovering the best genotype among a large number of candidates.

KEY MESSAGE: A new pre-breeding strategy based on an optimization algorithm is proposed and evaluated via simulations. This strategy can find superior genotypes with less phenotyping effort. Genomic prediction is a promising approach to search for superior genotypes among a large number of accessions in germplasm collections preserved in gene banks. When some accessions are phenotyped and genotyped, a prediction model can be built, and the genotypic values of the remaining accessions can be predicted from their marker genotypes. In this study, we focused on the application of genomic prediction to pre-breeding, and propose a novel strategy that would reduce the cost of phenotyping needed to discover better accessions. We regarded the exploration of superior genotypes with genomic prediction as an optimization problem, and introduced Bayesian optimization to solve it. Bayesian optimization, that samples unobserved inputs according to the expected improvement (EI) as a selection criterion, seemed to be beneficial in pre-breeding. The EI depends on the predicted distribution of genotypic values, whereas usual selection depends only on the point estimate. We simulated a search for the best genotype among candidate genotypes and showed that the EI-based strategy required fewer genotypes to identify the best genotype than the usual and random selection strategy. Therefore, Bayesian optimization can be useful for applying genomic prediction to pre-breeding and would reduce the number of phenotyped accessions needed to find the best accession among a large number of candidates.