Prospectus of Genomic Selection and Phenomics in Cereal, Legume and Oilseed Breeding Programs

Implementing multi-trait genomic selection to improve grain milling quality in oats (Avena sativa L.)

Oats (Avena sativa L.) provide unique nutritional benefits and contribute to sustainable agricultural systems. Breeding high-value oat varieties that meet milling industry standards is crucial for satisfying the demand for oat-based food products. Test weight, thins, and groat percentage are primary traits that define oat milling quality and the final price of food-grade oats. Conventional selection for milling quality is costly and burdensome. Multi-trait genomic selection (MTGS) combines information from genome-wide markers and secondary traits genetically correlated with primary traits to predict breeding values of primary traits on candidate breeding lines. MTGS can improve prediction accuracy and significantly accelerate the rate of genetic gain. In this study, we evaluated different MTGS models that used morphometric grain traits to improve prediction accuracy for primary grain quality traits within the constraints of a breeding program. We evaluated 558 breeding lines from the University of Illinois Oat Breeding Program across 2 years for primary milling traits, test weight, thins, and groat percentage, and secondary grain morphometric traits derived from kernel and groat images. Kernel morphometric traits were genetically correlated with test weight and thins percentage but were uncorrelated with groat percentage. For test weight and thins percentage, the MTGS model that included the kernel morphometric traits in both training and candidate sets outperformed single-trait models by 52% and 59%, respectively. In contrast, MTGS models for groat percentage were not significantly better than the single-trait model. We found that incorporating kernel morphometric traits can improve the genomic selection for test weight and thins percentage.

Are oilseeds a new alternative protein source for human nutrition?

This review focuses on the potential use, nutritional value and beneficial health effects of oilseeds as a source of food protein. The process of extracting oil from oilseeds produces a by-product that is rich in proteins and other valuable nutritional and bioactive components. This product is primarily used for animal feed. However, as the demand for proteins continues to rise, plant-based proteins have a real success in food applications. Among the different plant protein sources, oilseeds could be used as an alternative protein source for human diet. The data we have so far show that oilseeds present a protein content of up to 40% and a relatively well-balanced profile of amino acids with sulphur-containing amino acids. Nevertheless, they tend to be deficient in lysine and rich in anti-nutritional factors (ANFs), which therefore means they have lower anabolic potential than animal proteins. To enhance their nutritional value, oilseed proteins can be combined with other protein sources and subjected to processes such as dehulling, heating, soaking, germination or fermentation to reduce their ANFs and improve protein digestibility. Furthermore, due to their bioactive peptides, oilseeds can also bring health benefits, particularly in the prevention and treatment of diabetes, obesity and cardiovascular diseases. However, additional nutritional data are needed before oilseeds can be endorsed as a protein source for humans.

Smart reprograming of plants against salinity stress using modern biotechnological tools

Climate change gives rise to numerous environmental stresses, including soil salinity. Salinity/salt stress is the second biggest abiotic factor affecting agricultural productivity worldwide by damaging numerous physiological, biochemical, and molecular processes. In particular, salinity affects plant growth, development, and productivity. Salinity responses include modulation of ion homeostasis, antioxidant defense system induction, and biosynthesis of numerous phytohormones and osmoprotectants to protect plants from osmotic stress by decreasing ion toxicity and augmented reactive oxygen species scavenging. As most crop plants are sensitive to salinity, improving salt tolerance is crucial in sustaining global agricultural productivity. In response to salinity, plants trigger stress-related genes, proteins, and the accumulation of metabolites to cope with the adverse consequence of salinity. Therefore, this review presents an overview of salinity stress in crop plants. We highlight advances in modern biotechnological tools, such as omics (genomics, transcriptomics, proteomics, and metabolomics) approaches and different genome editing tools (ZFN, TALEN, and CRISPR/Cas system) for improving salinity tolerance in plants and accomplish the goal of "zero hunger," a worldwide sustainable development goal proposed by the FAO.

A simulation study comparing advanced marker-assisted selection with genomic selection in tree breeding programs

Advances in DNA sequencing technologies allow the sequencing of whole genomes of thousands of individuals and provide several million single nucleotide polymorphisms (SNPs) per individual. These data combined with precise and high-throughput phenotyping enable genome-wide association studies (GWAS) and the identification of SNPs underlying traits with complex genetic architectures. The identified causal SNPs and estimated allelic effects could then be used for advanced marker-assisted selection (MAS) in breeding programs. But could such MAS compete with the broadly used genomic selection (GS)? This question is of particular interest for the lengthy tree breeding strategies. Here, with our new software "SNPscan breeder," we simulated a simple tree breeding program and compared the impact of different selection criteria on genetic gain and inbreeding. Further, we assessed different genetic architectures and different levels of kinship among individuals of the breeding population. Interestingly, apart from progeny testing, GS using gBLUP performed best under almost all simulated scenarios. MAS based on GWAS results outperformed GS only if the allelic effects were estimated in large populations (ca. 10,000 individuals) of unrelated individuals. Notably, GWAS using 3,000 extreme phenotypes performed as good as the use of 10,000 phenotypes. GS increased inbreeding and thus reduced genetic diversity more strongly compared to progeny testing and GWAS-based selection. We discuss the practical implications for tree breeding programs. In conclusion, our analyses further support the potential of GS for forest tree breeding and improvement, although MAS may gain relevance with decreasing sequencing costs in the future.

Multi-omics assisted breeding for biotic stress resistance in soybean

Biotic stress is a critical factor limiting soybean growth and development. Soybean responses to biotic stresses such as insects, nematodes, fungal, bacterial, and viral pathogens are governed by complex regulatory and defense mechanisms. Next-generation sequencing has availed research techniques and strategies in genomics and post-genomics. This review summarizes the available information on marker resources, quantitative trait loci, and marker-trait associations involved in regulating biotic stress responses in soybean. We discuss the differential expression of related genes and proteins reported in different transcriptomics and proteomics studies and the role of signaling pathways and metabolites reported in metabolomic studies. Recent advances in omics technologies offer opportunities to reshape and improve biotic stress resistance in soybean by altering gene regulation and/or other regulatory networks. We suggest using 'integrated omics' to precisely understand how soybean responds to different biotic stresses. We also discuss the potential challenges of integrating multi-omics for the functional analysis of genes and their regulatory networks and the development of biotic stress-resistant cultivars. This review will help direct soybean breeding programs to develop resistance against different biotic stresses.

GWAS and genomic prediction for pre-harvest sprouting tolerance involving sprouting score and two other related traits in spring wheat

In wheat, a genome-wide association study (GWAS) and genomic prediction (GP) analysis were conducted for pre-harvest sprouting (PHS) tolerance and two of its related traits. For this purpose, an association panel of 190 accessions was phenotyped for PHS (using sprouting score), falling number, and grain color over two years and genotyped with 9904 DArTseq based SNP markers. GWAS for main-effect quantitative trait nucleotides (M-QTNs) using three different models (CMLM, SUPER, and FarmCPU) and epistatic QTNs (E-QTNs) using PLINK were performed. A total of 171 M-QTNs (CMLM, 47; SUPER, 70; FarmCPU, 54) for all three traits, and 15 E-QTNs involved in 20 first-order epistatic interactions were identified. Some of the above QTNs overlapped the previously reported QTLs, MTAs, and cloned genes, allowing delineating 26 PHS-responsive genomic regions that spread over 16 wheat chromosomes. As many as 20 definitive and stable QTNs were considered important for use in marker-assisted recurrent selection (MARS). The gene, TaPHS1, for PHS tolerance (PHST) associated with one of the QTNs was also validated using the KASP assay. Some of the M-QTNs were shown to have a key role in the abscisic acid pathway involved in PHST. Genomic prediction accuracies (based on the cross-validation approach) using three different models ranged from 0.41 to 0.55, which are comparable to the results of previous studies. In summary, the results of the present study improved our understanding of the genetic architecture of PHST and its related traits in wheat and provided novel genomic resources for wheat breeding based on MARS and GP.

Supplementary Information

The online version contains supplementary material available at 10.1007/s11032-023-01357-5.

Important wheat diseases in the US and their management in the 21st century

Wheat is a crop of historical significance, as it marks the turning point of human civilization 10,000 years ago with its domestication. Due to the rapid increase in population, wheat production needs to be increased by 50% by 2050 and this growth will be mainly based on yield increases, as there is strong competition for scarce productive arable land from other sectors. This increasing demand can be further achieved using sustainable approaches including integrated disease pest management, adaption to warmer climates, less use of water resources and increased frequency of abiotic stress tolerances. Out of 200 diseases of wheat, 50 cause economic losses and are widely distributed. Each year, about 20% of wheat is lost due to diseases. Some major wheat diseases are rusts, smut, tan spot, spot blotch, fusarium head blight, common root rot, septoria blotch, powdery mildew, blast, and several viral, nematode, and bacterial diseases. These diseases badly impact the yield and cause mortality of the plants. This review focuses on important diseases of the wheat present in the United States, with comprehensive information of causal organism, economic damage, symptoms and host range, favorable conditions, and disease management strategies. Furthermore, major genetic and breeding efforts to control and manage these diseases are discussed. A detailed description of all the QTLs, genes reported and cloned for these diseases are provided in this review. This study will be of utmost importance to wheat breeding programs throughout the world to breed for resistance under changing environmental conditions.

Multi-locus genome-wide association studies reveal genomic regions and putative candidate genes associated with leaf spot diseases in African groundnut (Arachis hypogaea L.) germplasm

Early leaf spot (ELS) and late leaf spot (LLS) diseases are the two most destructive groundnut diseases in Ghana resulting in ≤ 70% yield losses which is controlled largely by chemical method. To develop leaf spot resistant varieties, the present study was undertaken to identify single nucleotide polymorphism (SNP) markers and putative candidate genes underlying both ELS and LLS. In this study, six multi-locus models of genome-wide association study were conducted with the best linear unbiased predictor obtained from 294 African groundnut germplasm screened for ELS and LLS as well as image-based indices of leaf spot diseases severity in 2020 and 2021 and 8,772 high-quality SNPs from a 48 K SNP array Axiom platform. Ninety-seven SNPs associated with ELS, LLS and five image-based indices across the chromosomes in the 2 two sub-genomes. From these, twenty-nine unique SNPs were detected by at least two models for one or more traits across 16 chromosomes with explained phenotypic variation ranging from 0.01 - 62.76%, with exception of chromosome (Chr) 08 (Chr08), Chr10, Chr11, and Chr19. Seventeen potential candidate genes were predicted at ± 300 kbp of the stable/prominent SNP positions (12 and 5, down- and upstream, respectively). The results from this study provide a basis for understanding the genetic architecture of ELS and LLS diseases in African groundnut germplasm, and the associated SNPs and predicted candidate genes would be valuable for breeding leaf spot diseases resistant varieties upon further validation.

Recent advancement in OMICS approaches to enhance abiotic stress tolerance in legumes

The world is facing rapid climate change and a fast-growing global population. It is believed that the world population will be 9.7 billion in 2050. However, recent agriculture production is not enough to feed the current population of 7.9 billion people, which is causing a huge hunger problem. Therefore, feeding the 9.7 billion population in 2050 will be a huge target. Climate change is becoming a huge threat to global agricultural production, and it is expected to become the worst threat to it in the upcoming years. Keeping this in view, it is very important to breed climate-resilient plants. Legumes are considered an important pillar of the agriculture production system and a great source of high-quality protein, minerals, and vitamins. During the last two decades, advancements in OMICs technology revolutionized plant breeding and emerged as a crop-saving tool in wake of the climate change. Various OMICs approaches like Next-Generation sequencing (NGS), Transcriptomics, Proteomics, and Metabolomics have been used in legumes under abiotic stresses. The scientific community successfully utilized these platforms and investigated the Quantitative Trait Loci (QTL), linked markers through genome-wide association studies, and developed KASP markers that can be helpful for the marker-assisted breeding of legumes. Gene-editing techniques have been successfully proven for soybean, cowpea, chickpea, and model legumes such as Medicago truncatula and Lotus japonicus. A number of efforts have been made to perform gene editing in legumes. Moreover, the scientific community did a great job of identifying various genes involved in the metabolic pathways and utilizing the resulted information in the development of climate-resilient legume cultivars at a rapid pace. Keeping in view, this review highlights the contribution of OMICs approaches to abiotic stresses in legumes. We envisage that the presented information will be helpful for the scientific community to develop climate-resilient legume cultivars.

Genome-wide association studies of seedling quantitative trait loci against salt tolerance in wheat

Salinity is one of the significant factors in decreasing wheat yield and quality. To counter this, it is necessary to develop salt-tolerant wheat varieties through conventional and advanced molecular techniques. The current study identified quantitative trait loci in response to salt stress among worldwide landraces and improved varieties of wheat at the seedling stage. A total of 125 landraces and wheat varieties were subjected to salt treatment (50, 100, and 150 mM) with control. Morphological seedling traits, i.e., shoot length, root length, and fresh and dry shoot and root weights for salinity tolerance were observed to assess salt tolerance and genetic analysis using SNP data through DArT-seq. The results showed that, at the seedling stage, 150 mM NaCl treatment decreased shoot length, root length, and fresh and dry weights of the shoot and root. The root length and dry root weight were the most affected traits at the seedling stage. Effective 4417 SNPs encompassing all the chromosomes of the wheat genome with marker density, i.e., 37%, fall in genome B, genome D (32%), and genome A (31%). Five loci were found on four chromosomes 6B, 6D, 7A, and 7D, showing strong associations with the root length, fresh shoot weight, fresh root weight, and dry root weight at the p < 0.03 significance level. The positive correlation was found among all morphological traits under study.

CRISPR/Cas tool designs for multiplex genome editing and its applications in developing biotic and abiotic stress-resistant crop plants

Crop plants are prone to several yield-reducing biotic and abiotic stresses. The crop yield reductions due to these stresses need addressing to maintain an adequate balance between the increasing world population and food production to avoid food scarcities in the future. It is impossible to increase the area under food crops proportionately to meet the rising food demand. In such an adverse scenario overcoming the biotic and abiotic stresses through biotechnological interventions may serve as a boon to help meet the globe's food requirements. Under the current genomic era, the wide availability of genomic resources and genome editing technologies such as Transcription Activator-Like Effector Nucleases (TALENs), Zinc Finger Nucleases (ZFNs), and Clustered-Regularly Interspaced Palindromic Repeats/CRISPR-associated proteins (CRISPR/Cas) has widened the scope of overcoming these stresses for several food crops. These techniques have made gene editing more manageable and accessible with changes at the embryo level by adding or deleting DNA sequences of the target gene(s) from the genome. The CRISPR construct consists of a single guide RNA having complementarity with the nucleotide fragments of the target gene sequence, accompanied by a protospacer adjacent motif. The target sequence in the organism's genome is then cleaved by the Cas9 endonuclease for obtaining a desired trait of interest. The current review describes the components, mechanisms, and types of CRISPR/Cas techniques and how this technology has helped to functionally characterize genes associated with various biotic and abiotic stresses in a target organism. This review also summarizes the application of CRISPR/Cas technology targeting these stresses in crops through knocking down/out of associated genes.

Integrated Approach in Genomic Selection to Accelerate Genetic Gain in Sugarcane

Marker-assisted selection (MAS) has been widely used in the last few decades in plant breeding programs for the mapping and introgression of genes for economically important traits, which has enabled the development of a number of superior cultivars in different crops. In sugarcane, which is the most important source for sugar and bioethanol, marker development work was initiated long ago; however, marker-assisted breeding in sugarcane has been lagging, mainly due to its large complex genome, high levels of polyploidy and heterozygosity, varied number of chromosomes, and use of low/medium-density markers. Genomic selection (GS) is a proven technology in animal breeding and has recently been incorporated in plant breeding programs. GS is a potential tool for the rapid selection of superior genotypes and accelerating breeding cycle. However, its full potential could be realized by an integrated approach combining high-throughput phenotyping, genotyping, machine learning, and speed breeding with genomic selection. For better understanding of GS integration, we comprehensively discuss the concept of genetic gain through the breeder’s equation, GS methodology, prediction models, current status of GS in sugarcane, challenges of prediction accuracy, challenges of GS in sugarcane, integrated GS, high-throughput phenotyping (HTP), high-throughput genotyping (HTG), machine learning, and speed breeding followed by its prospective applications in sugarcane improvement.

Consensus genomic regions associated with multiple abiotic stress tolerance in wheat and implications for wheat breeding

In wheat, a meta-analysis was performed using previously identified QTLs associated with drought stress (DS), heat stress (HS), salinity stress (SS), water-logging stress (WS), pre-harvest sprouting (PHS), and aluminium stress (AS) which predicted a total of 134 meta-QTLs (MQTLs) that involved at least 28 consistent and stable MQTLs conferring tolerance to five or all six abiotic stresses under study. Seventy-six MQTLs out of the 132 physically anchored MQTLs were also verified with genome-wide association studies. Around 43% of MQTLs had genetic and physical confidence intervals of less than 1 cM and 5 Mb, respectively. Consequently, 539 genes were identified in some selected MQTLs providing tolerance to 5 or all 6 abiotic stresses. Comparative analysis of genes underlying MQTLs with four RNA-seq based transcriptomic datasets unravelled a total of 189 differentially expressed genes which also included at least 11 most promising candidate genes common among different datasets. The promoter analysis showed that the promoters of these genes include many stress responsiveness cis-regulatory elements, such as ARE, MBS, TC-rich repeats, As-1 element, STRE, LTR, WRE3, and WUN-motif among others. Further, some MQTLs also overlapped with as many as 34 known abiotic stress tolerance genes. In addition, numerous ortho-MQTLs among the wheat, maize, and rice genomes were discovered. These findings could help with fine mapping and gene cloning, as well as marker-assisted breeding for multiple abiotic stress tolerances in wheat.

Consensus genomic regions associated with multiple abiotic stress tolerance in wheat and implications for wheat breeding

In wheat, a meta-analysis was performed using previously identified QTLs associated with drought stress (DS), heat stress (HS), salinity stress (SS), water-logging stress (WS), pre-harvest sprouting (PHS), and aluminium stress (AS) which predicted a total of 134 meta-QTLs (MQTLs) that involved at least 28 consistent and stable MQTLs conferring tolerance to five or all six abiotic stresses under study. Seventy-six MQTLs out of the 132 physically anchored MQTLs were also verified with genome-wide association studies. Around 43% of MQTLs had genetic and physical confidence intervals of less than 1 cM and 5 Mb, respectively. Consequently, 539 genes were identified in some selected MQTLs providing tolerance to 5 or all 6 abiotic stresses. Comparative analysis of genes underlying MQTLs with four RNA-seq based transcriptomic datasets unravelled a total of 189 differentially expressed genes which also included at least 11 most promising candidate genes common among different datasets. The promoter analysis showed that the promoters of these genes include many stress responsiveness cis-regulatory elements, such as ARE, MBS, TC-rich repeats, As-1 element, STRE, LTR, WRE3, and WUN-motif among others. Further, some MQTLs also overlapped with as many as 34 known abiotic stress tolerance genes. In addition, numerous ortho-MQTLs among the wheat, maize, and rice genomes were discovered. These findings could help with fine mapping and gene cloning, as well as marker-assisted breeding for multiple abiotic stress tolerances in wheat.

Construction of Genetic Linkage Map and Mapping QTL Specific to Leaf Anthocyanin Colouration in Mapping Population 'Allahabad Safeda' × 'Purple Guava (Local)' of Guava (Psidium guajava L.)

In the present investigation, F1 hybrids were developed in guava (Psidium guajava L.) by crossing high leaf-anthocyanin reflective-index (ARI1) content cultivars purple guava (local) 'PG', 'CISH G-1' and low leaf-ARI1 content cultivar Seedless 'SL' with Allahabad Safeda 'AS'. On the basis of phenotypic observations, high ARI1 content was observed in the cross 'AS' × 'PG' (0.214). Further, an SSR-markers-based genetic linkage map was developed from a mapping population of 238 F1 individuals derived from cross 'AS' × 'PG'. The linkage map comprised 11 linkage groups (LGs), spanning 1601.7 cM with an average marker interval distance of 29.61 cM between adjacent markers. Five anthocyanin-content related gene-specific markers from apple were tested for parental polymorphism in the genotypes 'AS' and 'PG'. Subsequently, a marker, viz., 'MdMYB10F1', revealed a strong association with leaf anthocyanin content in the guava mapping population. QTL (qARI-6-1) on LG6 explains much of the variation (PVE = 11.51% with LOD = 4.67) in levels of leaf anthocyanin colouration. This is the first report of amplification/utilization of apple anthocyanin-related genes in guava. The genotypic data generated from the genetic map can be further exploited in future for the enrichment of linkage maps and for identification of complex quantitative trait loci (QTLs) governing economically important fruit quality traits in guava.

A Comprehensive Review of High Throughput Phenotyping and Machine Learning for Plant Stress Phenotyping

During the last decade, there has been rapid adoption of ground and aerial platforms with multiple sensors for phenotyping various biotic and abiotic stresses throughout the developmental stages of the crop plant. High throughput phenotyping (HTP) involves the application of these tools to phenotype the plants and can vary from ground-based imaging to aerial phenotyping to remote sensing. Adoption of these HTP tools has tried to reduce the phenotyping bottleneck in breeding programs and help to increase the pace of genetic gain. More specifically, several root phenotyping tools are discussed to study the plant's hidden half and an area long neglected. However, the use of these HTP technologies produces big data sets that impede the inference from those datasets. Machine learning and deep learning provide an alternative opportunity for the extraction of useful information for making conclusions. These are interdisciplinary approaches for data analysis using probability, statistics, classification, regression, decision theory, data visualization, and neural networks to relate information extracted with the phenotypes obtained. These techniques use feature extraction, identification, classification, and prediction criteria to identify pertinent data for use in plant breeding and pathology activities. This review focuses on the recent findings where machine learning and deep learning approaches have been used for plant stress phenotyping with data being collected using various HTP platforms. We have provided a comprehensive overview of different machine learning and deep learning tools available with their potential advantages and pitfalls. Overall, this review provides an avenue for studying various HTP platforms with particular emphasis on using the machine learning and deep learning tools for drawing legitimate conclusions. Finally, we propose the conceptual challenges being faced and provide insights on future perspectives for managing those issues.

Meta-analysis reveals consensus genomic regions associated with multiple disease resistance in wheat (Triticum aestivum L.)

In wheat, meta-QTLs (MQTLs) and candidate genes (CGs) were identified for multiple disease resistance (MDR). For this purpose, information was collected from 58 studies for mapping QTLs for resistance to one or more of the five diseases. As many as 493 QTLs were available from these studies, which were distributed in five diseases as follows: septoria tritici blotch (STB) 126 QTLs; septoria nodorum blotch (SNB), 103 QTLs; fusarium head blight (FHB), 184 QTLs; karnal bunt (KB), 66 QTLs; and loose smut (LS), 14 QTLs. Of these 493 QTLs, only 291 QTLs could be projected onto a consensus genetic map, giving 63 MQTLs. The CI of the MQTLs ranged from 0.04 to 15.31 cM with an average of 3.09 cM per MQTL. This is a ~ 4.39 fold reduction from the CI of QTLs, which ranged from 0 to 197.6 cM, with a mean of 13.57 cM. Of 63 MQTLs, 60 were anchored to the reference physical map of wheat (the physical interval of these MQTLs ranged from 0.30 to 726.01 Mb with an average of 74.09 Mb). Thirty-eight (38) of these MQTLs were verified using marker-trait associations (MTAs) derived from genome-wide association studies. As many as 874 CGs were also identified which were further investigated for differential expression using data from five transcriptome studies, resulting in 194 differentially expressed candidate genes (DECGs). Among the DECGs, 85 genes had functions previously reported to be associated with disease resistance. These results should prove useful for fine mapping and cloning of MDR genes and marker-assisted breeding.

Supplementary Information

The online version contains supplementary material available at 10.1007/s11032-022-01282-z.

Multi-Trait Multi-Environment Genomic Prediction for End-Use Quality Traits in Winter Wheat

Soft white wheat is a wheat class used in foreign and domestic markets to make various end products requiring specific quality attributes. Due to associated cost, time, and amount of seed needed, phenotyping for the end-use quality trait is delayed until later generations. Previously, we explored the potential of using genomic selection (GS) for selecting superior genotypes earlier in the breeding program. Breeders typically measure multiple traits across various locations, and it opens up the avenue for exploring multi-trait-based GS models. This study's main objective was to explore the potential of using multi-trait GS models for predicting seven different end-use quality traits using cross-validation, independent prediction, and across-location predictions in a wheat breeding program. The population used consisted of 666 soft white wheat genotypes planted for 5 years at two locations in Washington, United States. We optimized and compared the performances of four uni-trait- and multi-trait-based GS models, namely, Bayes B, genomic best linear unbiased prediction (GBLUP), multilayer perceptron (MLP), and random forests. The prediction accuracies for multi-trait GS models were 5.5 and 7.9% superior to uni-trait models for the within-environment and across-location predictions. Multi-trait machine and deep learning models performed superior to GBLUP and Bayes B for across-location predictions, but their advantages diminished when the genotype by environment component was included in the model. The highest improvement in prediction accuracy, that is, 35% was obtained for flour protein content with the multi-trait MLP model. This study showed the potential of using multi-trait-based GS models to enhance prediction accuracy by using information from previously phenotyped traits. It would assist in speeding up the breeding cycle time in a cost-friendly manner.