Using visual scores for genomic prediction of complex traits in breeding programs

KEY MESSAGE

An approach for handling visual scores with potential errors and subjectivity in scores was evaluated in simulated and blueberry recurrent selection breeding schemes to assist breeders in their decision-making. Most genomic prediction methods are based on assumptions of normality due to their simplicity and ease of implementation. However, in plant and animal breeding, continuous traits are often visually scored as categorical traits and analyzed as a Gaussian variable, thus violating the normality assumption, which could affect the prediction of breeding values and the estimation of genetic parameters. In this study, we examined the main challenges of visual scores for genomic prediction and genetic parameter estimation using mixed models, Bayesian, and machine learning methods. We evaluated these approaches using simulated and real breeding data sets. Our contribution in this study is a five-fold demonstration: (i) collecting data using an intermediate number of categories (1-3 and 1-5) is the best strategy, even considering errors associated with visual scores; (ii) Linear Mixed Models and Bayesian Linear Regression are robust to the normality violation, but marginal gains can be achieved when using Bayesian Ordinal Regression Models (BORM) and Random Forest Classification; (iii) genetic parameters are better estimated using BORM; (iv) our conclusions using simulated data are also applicable to real data in autotetraploid blueberry; and (v) a comparison of continuous and categorical phenotypes found that investing in the evaluation of 600-1000 categorical data points with low error, when it is not feasible to collect continuous phenotypes, is a strategy for improving predictive abilities. Our findings suggest the best approaches for effectively using visual scores traits to explore genetic information in breeding programs and highlight the importance of investing in the training of evaluator teams and in high-quality phenotyping.

Blueberry and cranberry pangenomes as a resource for future genetic studies and breeding efforts

Domestication of cranberry and blueberry began in the United States in the early 1800s and 1900s, respectively, and in part owing to their flavors and health-promoting benefits are now cultivated and consumed worldwide. The industry continues to face a wide variety of production challenges (e.g. disease pressures), as well as a demand for higher-yielding cultivars with improved fruit quality characteristics. Unfortunately, molecular tools to help guide breeding efforts for these species have been relatively limited compared with those for other high-value crops. Here, we describe the construction and analysis of the first pangenome for both blueberry and cranberry. Our analysis of these pangenomes revealed both crops exhibit great genetic diversity, including the presence-absence variation of 48.4% genes in highbush blueberry and 47.0% genes in cranberry. Auxiliary genes, those not shared by all cultivars, are significantly enriched with molecular functions associated with disease resistance and the biosynthesis of specialized metabolites, including compounds previously associated with improving fruit quality traits. The discovery of thousands of genes, not present in the previous reference genomes for blueberry and cranberry, will serve as the basis of future research and as potential targets for future breeding efforts. The pangenome, as a multiple-sequence alignment, as well as individual annotated genomes, are publicly available for analysis on the Genome Database for Vaccinium-a curated and integrated web-based relational database. Lastly, the core-gene predictions from the pangenomes will serve useful to develop a community genotyping platform to guide future molecular breeding efforts across the family.

Blueberry and cranberry pangenomes as a resource for future genetic studies and breeding efforts

Domestication of cranberry and blueberry began in the United States in the early 1800s and 1900s, respectively, and in part owing to their flavors and health-promoting benefits are now cultivated and consumed worldwide. The industry continues to face a wide variety of production challenges (e.g. disease pressures) as well as a demand for higher-yielding cultivars with improved fruit quality characteristics. Unfortunately, molecular tools to help guide breeding efforts for these species have been relatively limited compared with those for other high-value crops. Here, we describe the construction and analysis of the first pangenome for both blueberry and cranberry. Our analysis of these pangenomes revealed both crops exhibit great genetic diversity, including the presence-absence variation of 48.4% genes in highbush blueberry and 47.0% genes in cranberry. Auxiliary genes, those not shared by all cultivars, are significantly enriched with molecular functions associated with disease resistance and the biosynthesis of specialized metabolites, including compounds previously associated with improving fruit quality traits. The discovery of thousands of genes, not present in the previous reference genomes for blueberry and cranberry, will serve as the basis of future research and as potential targets for future breeding efforts. The pangenome, as a multiple-sequence alignment, as well as individual annotated genomes, are publicly available for analysis on the Genome Database for Vaccinium - a curated and integrated web-based relational database. Lastly, the core-gene predictions from the pangenomes will serve useful to develop a community genotyping platform to guide future molecular breeding efforts across the family.

Genomic prediction for canopy height and dry matter yield in alfalfa using family bulks

Genomic selection (GS) has proven to be an effective method to increase genetic gain rates and accelerate breeding cycles in many crop species. However, its implementation requires large investments to phenotype of the training population and for routine genotyping. Alfalfa (Medicago sativa L.) is one of the major cultivated forage legumes, showing high-quality nutritional value. Alfalfa breeding is usually carried out by phenotypic recurrent selection and is commonly done at the family level. The application of GS in alfalfa could be simplified and less costly by genotyping and phenotyping families in bulks. For this study, an alfalfa reference population composed of 142 full-sib and 35 half-sib families was bulk-genotyped using target enrichment sequencing and phenotyped for dry matter yield (DMY) and canopy height (CH) in Florida, USA. Genotyping of the family bulks with 17,707 targeted probes resulted in 114,945 single-nucleotide polymorphisms. The markers revealed a population structure that matched the mating design, and the linkage disequilibrium slowly decayed in this breeding population. After exploring multiple prediction scenarios, a strategy was proposed including data from multiple harvests and accounting for the G×E in the training population, which led to a higher predictive ability of up to 38 and 24% for DMY and CH, respectively. Although this study focused on the implementation of GS in alfalfa families, the bulk methodology and the prediction schemes used herein could guide future studies in alfalfa and other crops bred in bulks.

There and back again; historical perspective and future directions for Vaccinium breeding and research studies

The genus Vaccinium L. (Ericaceae) contains a wide diversity of culturally and economically important berry crop species. Consumer demand and scientific research in blueberry (Vaccinium spp.) and cranberry (Vaccinium macrocarpon) have increased worldwide over the crops' relatively short domestication history (~100 years). Other species, including bilberry (Vaccinium myrtillus), lingonberry (Vaccinium vitis-idaea), and ohelo berry (Vaccinium reticulatum) are largely still harvested from the wild but with crop improvement efforts underway. Here, we present a review article on these Vaccinium berry crops on topics that span taxonomy to genetics and genomics to breeding. We highlight the accomplishments made thus far for each of these crops, along their journey from the wild, and propose research areas and questions that will require investments by the community over the coming decades to guide future crop improvement efforts. New tools and resources are needed to underpin the development of superior cultivars that are not only more resilient to various environmental stresses and higher yielding, but also produce fruit that continue to meet a variety of consumer preferences, including fruit quality and health related traits.

Genomic Selection in an Outcrossing Autotetraploid Fruit Crop: Lessons From Blueberry Breeding

Blueberry (Vaccinium corymbosum and hybrids) is a specialty crop with expanding production and consumption worldwide. The blueberry breeding program at the University of Florida (UF) has greatly contributed to expanding production areas by developing low-chilling cultivars better adapted to subtropical and Mediterranean climates of the globe. The breeding program has historically focused on recurrent phenotypic selection. As an autopolyploid, outcrossing, perennial, long juvenile phase crop, blueberry breeding cycles are costly and time consuming, which results in low genetic gains per unit of time. Motivated by applying molecular markers for a more accurate selection in the early stages of breeding, we performed pioneering genomic selection studies and optimization for its implementation in the blueberry breeding program. We have also addressed some complexities of sequence-based genotyping and model parametrization for an autopolyploid crop, providing empirical contributions that can be extended to other polyploid species. We herein revisited some of our previous genomic selection studies and showed for the first time its application in an independent validation set. In this paper, our contribution is three-fold: (i) summarize previous results on the relevance of model parametrizations, such as diploid or polyploid methods, and inclusion of dominance effects; (ii) assess the importance of sequence depth of coverage and genotype dosage calling steps; (iii) demonstrate the real impact of genomic selection on leveraging breeding decisions by using an independent validation set. Altogether, we propose a strategy for using genomic selection in blueberry, with the potential to be applied to other polyploid species of a similar background.

Complete Chromosome-Scale Genome Sequence Resource for Sporisorium panici-leucophaei, the Causal Agent of Sourgrass Smut Disease

Here, we present the first complete chromosome-level genome assembly of the smut fungus strain Sporisorium panici-leucophaei SPL10A, the causal agent of the sourgrass (Digitaria insularis) smut disease. Combining Illumina paired-end and Nanopore long reads, we generated a final assembly composed of 23 chromosomes (22 nuclear and one mitochondrial) with 18,915,934 bp. Gene prediction accomplished using extrinsic evidence from the sugarcane smut fungus Sporisorium scitamineum originated a total of 6,402 protein-encoding genes. The secretome (388 proteins) and the effectorome repertoires (68 candidates) were also predicted, given their crucial roles in plant-pathogen interactions. The complete telomere-to-telomere chromosome sequences of this poorly studied fungus will provide a valuable resource for future comparative genomic studies among smuts to unravel their underlying pathogenicity mechanisms.[Formula: see text] Copyright © 2021 The Author(s). This is an open access article distributed under the CC BY-NC-ND 4.0 International license.

Genome-wide association of volatiles reveals candidate loci for blueberry flavor

Plants produce a range of volatile organic compounds (VOCs), some of which are perceived by the human olfactory system, contributing to a myriad flavors. Despite the importance of flavor for consumer preference, most plant breeding programs have neglected it, mainly because of the costs of phenotyping and the complexity of disentangling the role of VOCs in human perception. To develop molecular breeding tools aimed at improving fruit flavor, we carried out target genotyping of and VOC extraction from a blueberry population. Metabolite genome-wide association analysis was used to elucidate the genetic architecture, while predictive models were tested to prove that VOCs can be accurately predicted using genomic information. A historical sensory panel was considered to assess how the volatiles influenced consumers. By gathering genomics, metabolomics, and the sensory panel, we demonstrated that VOCs are controlled by a few major genomic regions, some of which harbor biosynthetic enzyme-coding genes; can be accurately predicted using molecular markers; and can enhance or decrease consumers' overall liking. Here we emphasized how the understanding of the genetic basis and the role of VOCs in consumer preference can assist breeders in developing more flavorful cultivars at a more inexpensive and accelerated pace.

How can a high-quality genome assembly help plant breeders?

The decreasing costs of next-generation sequencing and the improvements in de novo sequence assemblers have made it possible to obtain reference genomes for most eukaryotes, including minor crops such as the blueberry (Vaccinium corymbosum). Nevertheless, these genomes are at various levels of completeness and few have been anchored to chromosome scale and/or are haplotype-phased. We highlight the impact of a high-quality genome assembly for plant breeding and genetic research by showing how it affects our understanding of the genetic architecture of important traits and aids marker selection and candidate gene detection. We compared the results of genome-wide association studies and genomic selection that were already published using a blueberry draft genome as reference with the results using the recent released chromosome-scale and haplotype-phased blueberry genome. We believe that the benefits shown herein reinforce the importance of genome assembly projects for other non-model species.

High-Resolution Linkage Map and QTL Analyses of Fruit Firmness in Autotetraploid Blueberry

Blueberry (Vaccinium corymbosum and hybrids) is an autotetraploid crop whose commercial relevance has been growing steadily during the last 20 years. However, the ever-increasing cost of labor for hand-picking blueberry is one main constraint in competitive marketing of the fruit. Machine harvestability is, therefore, a key trait for the blueberry industry. Understanding the genetic architecture of traits related to machine harvestability through Quantitative Trait Loci (QTL) mapping is the first step toward implementation of molecular breeding for faster genetic gains. Despite recent advances in software development for autotetraploid genetic mapping, a high-resolution map is still not available for blueberry. In this study, we crafted a map for autotetraploid low-chill highbush blueberry containing 11,292 SNP markers and a total size of 1,953.97 cM (average density of 5.78 markers/cM). This map was subsequently used to perform QTL analyses in 2-year field trials for a trait crucial to machine harvesting: fruit firmness. Preliminary insights were also sought for single evaluations of firmness retention after cold storage, and fruit detachment force traits. Significant QTL peaks were identified for all the traits and overlapping QTL intervals were detected for firmness across the years. We found low-to-moderate QTL effects explaining the phenotypic variance, which suggest a quantitative nature of these traits. The QTL intervals were further speculated for putative gene repertoire. Altogether, our findings provide the basis for future fine-mapping and molecular breeding efforts for machine harvesting in blueberry.

Comparative Genomics of Smut Pathogens: Insights From Orphans and Positively Selected Genes Into Host Specialization

Host specialization is a key evolutionary process for the diversification and emergence of new pathogens. However, the molecular determinants of host range are poorly understood. Smut fungi are biotrophic pathogens that have distinct and narrow host ranges based on largely unknown genetic determinants. Hence, we aimed to expand comparative genomics analyses of smut fungi by including more species infecting different hosts and to define orphans and positively selected genes to gain further insights into the genetics basis of host specialization. We analyzed nine lineages of smut fungi isolated from eight crop and non-crop hosts: maize, barley, sugarcane, wheat, oats, Zizania latifolia (Manchurian rice), Echinochloa colona (a wild grass), and Persicaria sp. (a wild dicot plant). We assembled two new genomes: Ustilago hordei (strain Uhor01) isolated from oats and U. tritici (strain CBS 119.19) isolated from wheat. The smut genomes were of small sizes, ranging from 18.38 to 24.63 Mb. U. hordei species experienced genome expansions due to the proliferation of transposable elements and the amount of these elements varied among the two strains. Phylogenetic analysis confirmed that Ustilago is not a monophyletic genus and, furthermore, detected misclassification of the U. tritici specimen. The comparison between smut pathogens of crop and non-crop hosts did not reveal distinct signatures, suggesting that host domestication did not play a dominant role in shaping the evolution of smuts. We found that host specialization in smut fungi likely has a complex genetic basis: different functional categories were enriched in orphans and lineage-specific selected genes. The diversification and gain/loss of effector genes are probably the most important determinants of host specificity.

Using Population and Comparative Genomics to Understand the Genetic Basis of Effector-Driven Fungal Pathogen Evolution

Epidemics caused by fungal plant pathogens pose a major threat to agro-ecosystems and impact global food security. High-throughput sequencing enabled major advances in understanding how pathogens cause disease on crops. Hundreds of fungal genomes are now available and analyzing these genomes highlighted the key role of effector genes in disease. Effectors are small secreted proteins that enhance infection by manipulating host metabolism. Fungal genomes carry 100s of putative effector genes, but the lack of homology among effector genes, even for closely related species, challenges evolutionary and functional analyses. Furthermore, effector genes are often found in rapidly evolving chromosome compartments which are difficult to assemble. We review how population and comparative genomics toolsets can be combined to address these challenges. We highlight studies that associated genome-scale polymorphisms with pathogen lifestyles and adaptation to different environments. We show how genome-wide association studies can be used to identify effectors and other pathogenicity-related genes underlying rapid adaptation. We also discuss how the compartmentalization of fungal genomes into core and accessory regions shapes the evolution of effector genes. We argue that an understanding of genome evolution provides important insight into the trajectory of host-pathogen co-evolution.

Molecular variability and genetic relationship among Brazilian strains of the sugarcane smut fungus

Sporisorium scitamineum is the fungus that causes sugarcane smut disease. Despite of the importance of sugarcane for Brazilian agribusiness and the persistence of the pathogen in most cropping areas, genetic variation studies are still missing for Brazilian isolates. In this study, sets of isolates were analyzed using two molecular markers (AFLP and telRFLP) and ITS sequencing. Twenty-two whips were collected from symptomatic plants in cultivated sugarcane fields of Brazil. A total of 41 haploid strains of compatible mating types were selected from individual teliospores and used for molecular genetic analyses. telRFLP and ITS analyses were expanded to six Argentine isolates, where the sugarcane smut was first recorded in America. Genetic relationship among strains suggests the human-mediated dispersal of S. scitamineum within the Brazilian territory and between the two neighboring countries. Two genetically distinct groups were defined by the combined analysis of AFLP and telRFLP. The opposite mating-type strains derived from single teliospores were clustered together into these main groups, but had not always identical haplotypes. telRFLP markers analyzed over two generations of selfing and controlled outcrossing confirmed the potential for emergence of new variants and occurrence of recombination, which are relevant events for evolution of virulence and environmental adaptation.