Genomic selection, the application of genomic prediction (GP) models to select candidate individuals, has significantly advanced in the past two decades, effectively accelerating genetic gains in plant breeding. This article provides a holistic overview of the key factors that have influenced GP in plant breeding during this period. We delved into the pivotal roles of training population size and genetic diversity, and their relationship with the breeding population, in determining genomic prediction accuracy. Special emphasis was placed on optimizing training population size. We explored its benefits and the associated diminishing returns beyond an optimum size. This was done while considering the balance between resource allocation and maximizing prediction accuracy through current optimization algorithms. The density and distribution of single nucleotide polymorphisms (SNPs), level of linkage disequilibrium, genetic complexity, trait heritability, statistical machine learning methods and non-additive effects are the other vital factors. Using wheat, maize and potato as examples, we summarize the effect of these factors on the accuracy of GP for various traits. The search for high accuracy in GP - theoretically reaching one when using the Pearson's correlation as a metric - is an active research area yet far from optimal for various traits. We hypothesize that with ultra-high sizes of genotypic and phenotypic datasets, effective training population optimization methods, and support from other omics approaches (transcriptomics, metabolomics and proteomics) coupled with deep learning algorithms could overcome the boundaries of the current limitations to achieve the highest possible prediction accuracy, making GS an effective tool in plant breeding.

Septoria tritici blotch (STB) caused by the fungal pathogen Zymoseptoria tritici and powdery mildew (PM) caused by Blumeria graminis f.sp tritici (Bgt) are among the forefront foliar diseases of wheat that lead to a significant loss of grain yield and quality. Resistance breeding aimed at developing varieties with inherent resistance to STB and PM diseases has been the most sustainable and environment-friendly approach. In this study, 175 winter wheat landraces and historical cultivars originated from the Nordic region were evaluated for adult-plant resistance (APR) to STB and PM in Denmark, Estonia, Lithuania, and Sweden. Genome-wide association study (GWAS) and genomic prediction (GP) were performed based on the adult-plant response to STB and PM in field conditions using 7,401 single-nucleotide polymorphism (SNP) markers generated by 20K SNP chip. Genotype-by-environment interaction was significant for both disease scores. GWAS detected stable and environment-specific quantitative trait locis (QTLs) on chromosomes 1A, 1B, 1D, 2B, 3B, 4A, 5A, 6A, and 6B for STB and 2A, 2D, 3A, 4B, 5A, 6B, 7A, and 7B for PM adult-plant disease resistance. GP accuracy was improved when assisted with QTL from GWAS as a fixed effect. The GWAS-assisted GP accuracy ranged within 0.53–0.75 and 0.36–0.83 for STB and PM, respectively, across the tested environments. This study highlights that landraces and historical cultivars are a valuable source of APR to STB and PM. Such germplasm could be used to identify and introgress novel resistance genes to modern breeding lines.

Background: Ethiopia has been considered as a center of diversity and the second possible center of domestication of durum wheat. Genetic diversity and population structure analysis in the existing Ethiopian durum wheat germplasm have enormous importance in enhancing breeding effort and for sustainable conservation. Hence, 192 Ethiopian durum wheat accessions comprising 167 landraces collected from major wheat-growing areas of the country and 25 improved varieties released from Debre Zeit and Sinana Agricultural Research Centers, Ethiopia in different years (1994–2010) were assembled for the current study.Results: The panel was genotyped with a High-density 90 K wheat SNP array by Illumina and generated 15,338 polymorphic SNPs that were used to analyze the genetic diversity and to estimate the population structure. Varied values of genetic diversity indices were scored across chromosomes and genomes. Genome-wide mean values of Nei’s gene diversity (0.246) and polymorphism information content (0.203) were recorded signifying the presence of high genetic diversity within this collection. Minor allele frequency of the genome varied with a range of 0.005 to 0.5 scoring a mean value of 0.175. Improved varieties clustered separately to landraces in population structure analysis resulted from STRUCTURE, PCA and neighbor joining tree. Landraces clustering was irrespective of their geographical origin signifying the presence of higher admixture that could arise due to the existence of historical exchanges of seeds through informal seed system involving regional and countrywide farming communities in Ethiopia.Conclusions: Sustainable utilization and conservation of this rich Ethiopian durum wheat genetic resource is an irreplaceable means to cope up from the recurrent climate changes and biotic stresses happening widely and thereby able to keep meeting the demand of durum productivity for the ever-growing human population.

Background: Genetic improvement of root system architecture is essential to improve water and nutrient use efficiency of crops or to boost their productivity under stress or non-optimal soil conditions. One hundred ninety-two Ethiopian durum wheat accessions comprising 167 historical landraces and 25 modern cultivars were assembled for GWAS analysis to identify QTLs for root system architecture (RSA) traits and genotyped with a high-density 90 K wheat SNP array by Illumina.Results: Using a non-roll, paper-based root phenotyping platform, a total of 2880 seedlings and 14,947 seminal roots were measured at the three-leaf stage to collect data for total root length (TRL), total root number (TRN), root growth angle (RGA), average root length (ARL), bulk root dry weight (RDW), individual root dry weight (IRW), bulk shoot dry weight (SDW), presence of six seminal roots per seedling (RT6) and root shoot ratio (RSR). Analysis of variance revealed highly significant differences between accessions for all RSA traits. Four major (− log₁₀P≥ 4) and 34 nominal (− log₁₀P≥ 3) QTLs were identified and grouped in 16 RSA QTL clusters across chromosomes. A higher number of significant RSA QTL were identified on chromosome 4B particularly for root vigor traits (root length, number and/or weight).Conclusions: After projecting the identified QTLs on to a high-density tetraploid consensus map along with previously reported RSA QTL in both durum and bread wheat, fourteen nominal QTLs were found to be novel and could potentially be used to tailor RSA in elite lines. The major RGA QTLs on chromosome 6AL detected in the current study and reported in previous studies is a good candidate for cloning the causative underlining sequence and identifying the beneficial haplotypes able to positively affect yield under water- or nutrient-limited conditions.

Identification and exploration of the genetic architecture of traits related to yield, quality, and drought and heat tolerance is important for yield and quality improvement of wheat through marker-assisted selection. One hundred and ninety-two spring wheat genotypes were tested at two heat-stress locations in Sudan (Wad Medani and Dongula), a drought stress site in Morocco (Marchouch) and a site with high yield potential in Egypt (Sids) in replicated trials during the 2015–2016 and 2016–2017 cropping seasons. A total of 10,577 single nucleotide polymorphism markers identified from the 15 K wheat SNP assay were used in a genome-wide association (GWA) study and genomic prediction for 16 phenotypic traits related to yield, quality and drought and heat tolerance. Significant marker-trait associations were detected across GWAS models for all traits. Most detected marker-trait associations (MTAs) were environment-specific, signifying the presence of high quantitative trait loci-by-environment (QTL x E) interaction. Chromosome arm 5AL had significant multi-model MTAs for grain yield and yield-related traits at the heat-stress locations. Highly significant QTLs were detected on chromosome 2D for waxiness. Homoeologous group 2 and 6 chromosomes were with significant MTAs for grain protein content, gluten content, alveograph strength and Zeleny sedimentation test while chromosome arm 3BL was significant for both Z and W traits. Genomic prediction analysis with ridge regression-best linear unbiased prediction model estimated the breeding values of the studied traits with prediction accuracies ranging from 0.16 for leaf rolling to 0.72 for peduncle length. The identified QTLs could be targeted for marker-assisted selection or further studies aimed at fine mapping and cloning the causative genes and detecting favorable haplotypes with positive effects for agronomic, physiological or quality-related traits.