Harness the power of genomic selection and the potential of germplasm in crop breeding for global food security in the era with rapid climate change

GWAS analysis reveals the genetic basis of blast resistance associated with heading date in rice

Rice blast is a destructive fungal disease affecting rice plants at various growth stages, significantly threatening global yield stability. Development of resistant rice cultivars stands as a practical means of disease control. Generally, association mapping with a diversity panel powerfully identifies new alleles controlling trait of interest. On the other hand, utilization of a breeding panel has its advantage that can be directly applied in a breeding program. In this study, we conducted a genome-wide association study (GWAS) for blast resistance using 296 commercial rice cultivars with low population structure but large phenotypic diversity. We attempt to answer the genetic basis behind rice blast resistance among early maturing cultivars by subdividing the population based on its Heading date 1 (Hd1) functionality. Subpopulation-specific GWAS using the mixed linear model (MLM) based on blast nursery screening conducted in three years revealed a total of 26 significant signals, including three nucleotide-binding site leucine-rich repeat (NBS-LRR) genes (Os06g0286500, Os06g0286700, and Os06g0287500) located at Piz locus on chromosome 6, and one at the Pi-ta locus (Os12g0281300) on chromosome 12. Haplotype analysis revealed blast resistance associated with Piz locus was exclusively specific to Type 14 hd1 among japonica rice. Our findings provide valuable insights for breeding blast resistant rice and highlight the applicability of our elite cultivar panel to detect superior alleles associated with important agronomic traits.

Genetic gains in IRRI's rice salinity breeding and elite panel development as a future breeding resource

KEY MESSAGE

Estimating genetic gains and formulating a future salinity elite breeding panel for rice pave the way for developing better high-yielding salinity tolerant lines with enhanced genetic gains. Genetic gain is a crucial parameter to check the breeding program's success and help optimize future breeding strategies for enhanced genetic gains. To estimate the genetic gains in IRRI's salinity breeding program and identify the best genotypes based on high breeding values for grain yield (kg/ha), we analyzed the historical data from the trials conducted in the IRRI, Philippines and Bangladesh. A two-stage mixed-model approach accounting for experimental design factors and a relationship matrix was fitted to obtain the breeding values for grain yield and estimate genetic trends. A positive genetic trend of 0.1% per annum with a yield advantage of 1.52 kg/ha was observed in IRRI, Philippines. In Bangladesh, we observed a genetic gain of 0.31% per annum with a yield advantage of 14.02 kg/ha. In the released varieties, we observed a genetic gain of 0.12% per annum with a 2.2 kg/ha/year yield advantage in the IRRI, Philippines. For the Bangladesh dataset, a genetic gain of 0.14% per annum with a yield advantage of 5.9 kg/ha/year was observed in the released varieties. Based on breeding values for grain yield, a core set of the top 145 genotypes with higher breeding values of > 2400 kg/ha in the IRRI, Philippines, and > 3500 kg/ha in Bangladesh with a reliability of > 0.4 were selected to develop the elite breeding panel. Conclusively, a recurrent selection breeding strategy integrated with novel technologies like genomic selection and speed breeding is highly required to achieve higher genetic gains in IRRI's salinity breeding programs.

Genomics empowering conservation action and improvement of celery in the face of climate change

MAIN CONCLUSION

Integration of genomic approaches like whole genome sequencing, functional genomics, evolutionary genomics, and CRISPR/Cas9-based genome editing has accelerated the improvement of crop plants including leafy vegetables like celery in the face of climate change. The anthropogenic climate change is a real peril to the existence of life forms on our planet, including human and plant life. Climate change is predicted to be a significant threat to biodiversity and food security in the coming decades and is rapidly transforming global farming systems. To avoid the ghastly future in the face of climate change, the elucidation of shifts in the geographical range of plant species, species adaptation, and evolution is necessary for plant scientists to develop climate-resilient strategies. In the post-genomics era, the increasing availability of genomic resources and integration of multifaceted genomics elements is empowering biodiversity conservation action, restoration efforts, and identification of genomic regions adaptive to climate change. Genomics has accelerated the true characterization of crop wild relatives, genomic variations, and the development of climate-resilient varieties to ensure food security for 10 billion people by 2050. In this review, we have summarized the applications of multifaceted genomic tools, like conservation genomics, whole genome sequencing, functional genomics, genome editing, pangenomics, in the conservation and adaptation of plant species with a focus on celery, an aromatic and medicinal Apiaceae vegetable. We focus on how conservation scientists can utilize genomics and genomic data in conservation and improvement.

Drought response of water-conserving and non-conserving spring barley cultivars

Introduction

Breeding barley cultivars adapted to drought requires in-depth knowledge on physiological drought responses.

Methods

We used a high-throughput functional phenotyping platform to examine the response of four high-yielding European spring barley cultivars to a standardized drought treatment imposed around flowering.

Results

Cv. Chanell showed a non-conserving water-use behavior with high transpiration and maximum productivity under well-watered conditions but rapid transpiration decrease under drought. The poor recovery upon re-irrigation translated to large yield losses. Cv. Baronesse showed the most water-conserving behavior, with the lowest pre-drought transpiration and the most gradual transpiration reduction under drought. Its good recovery (resilience) prevented large yield losses. Cv. Formula was less conserving than cv. Baronesse and produced low yet stable yields. Cv. RGT's dynamic water use with high transpiration under ample water supply and moderate transpiration decrease under drought combined with high resilience secured the highest and most stable yields.

Discussion

Such a dynamic water-use behavior combined with higher drought resilience and favorable root traits could potentially create an ideotype for intermediate drought. Prospective studies will examine these results in field experiments and will use the newly gained understanding on water use in barley to improve process descriptions in crop simulation models to support crop model-aided ideotype design.

Plant breeding for harmony between sustainable agriculture, the environment, and global food security: an era of genomics-assisted breeding

MAIN CONCLUSION

Genomics-assisted breeding represents a crucial frontier in enhancing the balance between sustainable agriculture, environmental preservation, and global food security. Its precision and efficiency hold the promise of developing resilient crops, reducing resource utilization, and safeguarding biodiversity, ultimately fostering a more sustainable and secure food production system. Agriculture has been seriously threatened over the last 40 years by climate changes that menace global nutrition and food security. Changes in environmental factors like drought, salt concentration, heavy rainfalls, and extremely low or high temperatures can have a detrimental effects on plant development, growth, and yield. Extreme poverty and increasing food demand necessitate the need to break the existing production barriers in several crops. The first decade of twenty-first century marks the rapid development in the discovery of new plant breeding technologies. In contrast, in the second decade, the focus turned to extracting information from massive genomic frameworks, speculating gene-to-phenotype associations, and producing resilient crops. In this review, we will encompass the causes, effects of abiotic stresses and how they can be addressed using plant breeding technologies. Both conventional and modern breeding technologies will be highlighted. Moreover, the challenges like the commercialization of biotechnological products faced by proponents and developers will also be accentuated. The crux of this review is to mention the available breeding technologies that can deliver crops with high nutrition and climate resilience for sustainable agriculture.

Optimization of Multi-Generation Multi-location Genomic Prediction Models for Recurrent Genomic Selection in an Upland Rice Population

Genomic selection is a worthy breeding method to improve genetic gain in recurrent selection breeding schemes. The integration of multi-generation and multi-location information could significantly improve genomic prediction models in the context of shuttle breeding. The Cirad-CIAT upland rice breeding program applies recurrent genomic selection and seeks to optimize the scheme to increase genetic gain while reducing phenotyping efforts. We used a synthetic population (PCT27) of which S0 plants were all genotyped and advanced by selfing and bulk seed harvest to the S0:2, S0:3, and S0:4 generations. The PCT27 was then divided into two sets. The S0:2 and S0:3 progenies for PCT27A and the S0:4 progenies for PCT27B were phenotyped in two locations: Santa Rosa the target selection location, within the upland rice growing area, and Palmira, the surrogate location, far from the upland rice growing area but easier for experimentation. While the calibration used either one of the two sets phenotyped in one or two locations, the validation population was only the PCT27B phenotyped in Santa Rosa. Five scenarios of genomic prediction and 24 models were performed and compared. Training the prediction model with the PCT27B phenotyped in Santa Rosa resulted in predictive abilities ranging from 0.19 for grain zinc concentration to 0.30 for grain yield. Expanding the training set with the inclusion of the PCT27A resulted in greater predictive abilities for all traits but grain yield, with increases from 5% for plant height to 61% for grain zinc concentration. Models with the PCT27B phenotyped in two locations resulted in higher prediction accuracy when the models assumed no genotype-by-environment (G × E) interaction for flowering (0.38) and grain zinc concentration (0.27). For plant height, the model assuming a single G × E variance provided higher accuracy (0.28). The gain in predictive ability for grain yield was the greatest (0.25) when environment-specific variance deviation effect for G × E was considered. While the best scenario was specific to each trait, the results indicated that the gain in predictive ability provided by the multi-location and multi-generation calibration was low. Yet, this approach could lead to increased selection intensity, acceleration of the breeding cycle, and a sizable economic advantage for the program.

Plant transcriptional memory and associated mechanism of abiotic stress tolerance

Plants face various adverse environmental conditions, particularly with the ongoing changes in global climate, which drastically affect the growth, development and productivity of crops. To cope with these stresses, plants have evolved complex mechanisms, and one of the crucial ways is to develop transcriptional memories from stress exposure. This induced learning enables plants to better and more strongly restart the response and adaptation mechanism to stress when similar or dissimilar stresses reoccur. Understanding the molecular mechanism behind plant transcriptional memory of stress can provide a theoretical basis for breeding stress-tolerant crops with resilience to future climates. Here we review the recent research progress on the transcriptional memory of plants under various stresses and the applications of underlying mechanisms for sustainable agricultural production. We propose that a thorough understanding of plant transcriptional memory is crucial for both agronomic management and resistant breeding, and thus may help to improve agricultural yield and quality under changing climatic conditions.

The Era of Plant Breeding: Conventional Breeding to Genomics-assisted Breeding for Crop Improvement

Plant breeding has made a significant contribution to increasing agricultural production. Conventional breeding based on phenotypic selection is not effective for crop improvement. Because phenotype is considerably influenced by environmental factors, which will affect the selection of breeding materials for crop improvement. The past two decades have seen tremendous progress in plant breeding research. Especially the availability of high-throughput molecular markers followed by genomic-assisted approaches significantly contributed to advancing plant breeding. Integration of speed breeding with genomic and phenomic facilities allowed rapid quantitative trait loci (QTL)/gene identifications and ultimately accelerated crop improvement programs. The advances in sequencing technology helps to understand the genome organization of many crops and helped with genomic selection in crop breeding. Plant breeding has gradually changed from phenotype-to-genotype-based to genotype-to-phenotype-based selection. High-throughput phenomic platforms have played a significant role in the modern breeding program and are considered an essential part of precision breeding. In this review, we discuss the rapid advance in plant breeding technology for efficient crop improvements and provide details on various approaches/platforms that are helpful for crop improvement. This review will help researchers understand the recent developments in crop breeding and improvements.

A pangenome analysis pipeline provides insights into functional gene identification in rice

BACKGROUND

A pangenome aims to capture the complete genetic diversity within a species and reduce bias in genetic analysis inherent in using a single reference genome. However, the current linear format of most plant pangenomes limits the presentation of position information for novel sequences. Graph pangenomes have been developed to overcome this limitation. However, bioinformatics analysis tools for graph format genomes are lacking.

RESULTS

To overcome this problem, we develop a novel strategy for pangenome construction and a downstream pangenome analysis pipeline (PSVCP) that captures genetic variants' position information while maintaining a linearized layout. Using PSVCP, we construct a high-quality rice pangenome using 12 representative rice genomes and analyze an international rice panel with 413 diverse accessions using the pangenome as the reference. We show that PSVCP successfully identifies causal structural variations for rice grain weight and plant height. Our results provide insights into rice population structure and genomic diversity. We characterize a new locus (qPH8-1) associated with plant height on chromosome 8 undetected by the SNP-based genome-wide association study (GWAS).

CONCLUSIONS

Our results demonstrate that the pangenome constructed by our pipeline combined with a presence and absence variation-based GWAS can provide additional power for genomic and genetic analysis. The pangenome constructed in this study and the associated genome sequence and genetic variants data provide valuable genomic resources for rice genomics research and improvement in future.

DNNGP, a deep neural network-based method for genomic prediction using multi-omics data in plants

Genomic prediction is an effective way to accelerate the rate of agronomic trait improvement in plants. Traditional methods typically use linear regression models with clear assumptions; such methods are unable to capture the complex relationships between genotypes and phenotypes. Non-linear models (e.g., deep neural networks) have been proposed as a superior alternative to linear models because they can capture complex non-additive effects. Here we introduce a deep learning (DL) method, deep neural network genomic prediction (DNNGP), for integration of multi-omics data in plants. We trained DNNGP on four datasets and compared its performance with methods built with five classic models: genomic best linear unbiased prediction (GBLUP); two methods based on a machine learning (ML) framework, light gradient boosting machine (LightGBM) and support vector regression (SVR); and two methods based on a DL framework, deep learning genomic selection (DeepGS) and deep learning genome-wide association study (DLGWAS). DNNGP is novel in five ways. First, it can be applied to a variety of omics data to predict phenotypes. Second, the multilayered hierarchical structure of DNNGP dynamically learns features from raw data, avoiding overfitting and improving the convergence rate using a batch normalization layer and early stopping and rectified linear activation (rectified linear unit) functions. Third, when small datasets were used, DNNGP produced results that are competitive with results from the other five methods, showing greater prediction accuracy than the other methods when large-scale breeding data were used. Fourth, the computation time required by DNNGP was comparable with that of commonly used methods, up to 10 times faster than DeepGS. Fifth, hyperparameters can easily be batch tuned on a local machine. Compared with GBLUP, LightGBM, SVR, DeepGS and DLGWAS, DNNGP is superior to these existing widely used genomic selection (GS) methods. Moreover, DNNGP can generate robust assessments from diverse datasets, including omics data, and quickly incorporate complex and large datasets into usable models, making it a promising and practical approach for straightforward integration into existing GS platforms.

Phenotype based clustering, and diversity of common bean genotypes in seed iron concentration and cooking time

Common bean is the world's most important directly consumed legume food crop that is popular for calories, protein and micronutrients. It is a staple food in sub-Saharan Africa, and a significant source of iron for anemic people. However, several pests, soil and weather challenges still impede its production. Long cooking time, and high phytic acid and polyphenols that influence bioavailable iron also limit the health benefits. To inform population improvement strategies and selection decisions for resilient fast cooking and iron biofortified beans, the study determined diversity and population structure within 427 breeding lines, varieties, or landraces mostly from Alliance Uganda and Columbia. The genotypes were evaluated for days to flowering and physiological maturity, yield, seed iron (FESEED) and zinc (ZNSEED) and cooking time (COOKT). Data for all traits showed significant (P≤0.001) differences among the genotypes. Repeatability was moderate to high for most traits. Performance ranged from 52 to 87 ppm (FESEED), 23-38 ppm (ZNSEED), 36-361 minutes (COOKT), and 397-1299 kg/ha (yield). Minimal differences existed between the gene pools in the mean performance except in yield, where Mesoamerican beans were better by 117 kg/ha. The genotypes exhibited high genetic diversity and thus have a high potential for use in plant breeding. Improvement of FESEED and ZNSEED, COOKT and yield performance within some markets such as red and small white beans is possible. Hybridization across market classes especially for yellow beans is essential but this could be avoided by adding other elite lines to the population. Superior yielding and fast cooking, yellow and large white beans were specifically lacking. Adding Fe dense elite lines to the population is also recommended. The population was clustered into three groups that could be considered for specific breeding targets based on trait correlations.

Perspective for genomic-enabled prediction against black sigatoka disease and drought stress in polyploid species

Genomic selection (GS) in plant breeding is explored as a promising tool to solve the problems related to the biotic and abiotic threats. Polyploid plants like bananas (Musa spp.) face the problem of drought and black sigatoka disease (BSD) that restrict their production. The conventional plant breeding is experiencing difficulties, particularly phenotyping costs and long generation interval. To overcome these difficulties, GS in plant breeding is explored as an alternative with a great potential for reducing costs and time in selection process. So far, GS does not have the same success in polyploid plants as with diploid plants because of the complexity of their genome. In this review, we present the main constraints to the application of GS in polyploid plants and the prospects for overcoming these constraints. Particular emphasis is placed on breeding for BSD and drought-two major threats to banana production-used in this review as a model of polyploid plant. It emerges that the difficulty in obtaining markers of good quality in polyploids is the first challenge of GS on polyploid plants, because the main tools used were developed for diploid species. In addition to that, there is a big challenge of mastering genetic interactions such as dominance and epistasis effects as well as the genotype by environment interaction, which are very common in polyploid plants. To get around these challenges, we have presented bioinformatics tools, as well as artificial intelligence approaches, including machine learning. Furthermore, a scheme for applying GS to banana for BSD and drought has been proposed. This review is of paramount impact for breeding programs that seek to reduce the selection cycle of polyploids despite the complexity of their genome.

Multi-Omics Approaches and Resources for Systems-Level Gene Function Prediction in the Plant Kingdom

In higher plants, the complexity of a system and the components within and among species are rapidly dissected by omics technologies. Multi-omics datasets are integrated to infer and enable a comprehensive understanding of the life processes of organisms of interest. Further, growing open-source datasets coupled with the emergence of high-performance computing and development of computational tools for biological sciences have assisted in silico functional prediction of unknown genes, proteins and metabolites, otherwise known as uncharacterized. The systems biology approach includes data collection and filtration, system modelling, experimentation and the establishment of new hypotheses for experimental validation. Informatics technologies add meaningful sense to the output generated by complex bioinformatics algorithms, which are now freely available in a user-friendly graphical user interface. These resources accentuate gene function prediction at a relatively minimal cost and effort. Herein, we present a comprehensive view of relevant approaches available for system-level gene function prediction in the plant kingdom. Together, the most recent applications and sought-after principles for gene mining are discussed to benefit the plant research community. A realistic tabulation of plant genomic resources is included for a less laborious and accurate candidate gene discovery in basic plant research and improvement strategies.

Priming crops for the future: rewiring stress memory

The agricultural sector must produce resilient and climate-smart crops to meet the increasing needs of global food production. Recent advancements in elucidating the mechanistic basis of plant stress memory have provided new opportunities for crop improvement. Stress memory-coordinated changes at the organismal, cellular, and various omics levels prepare plants to be more responsive to reoccurring stress within or across generation(s). The exposure to a primary stress, or stress priming, can also elicit a beneficial impact when encountering a secondary abiotic or biotic stress through the convergence of synergistic signalling pathways, referred to as cross-stress tolerance. 'Rewired plants' with stress memory provide a new means to stimulate adaptable stress responses, safeguard crop reproduction, and engineer climate-smart crops for the future.

Indian Wheat Genomics Initiative for Harnessing the Potential of Wheat Germplasm Resources for Breeding Disease-Resistant, Nutrient-Dense, and Climate-Resilient Cultivars

Wheat is one of the major staple cereal food crops in India. However, most of the wheat-growing areas experience several biotic and abiotic stresses, resulting in poor quality grains and reduced yield. To ensure food security for the growing population in India, there is a compelling need to explore the untapped genetic diversity available in gene banks for the development of stress-resistant/tolerant cultivars. The improvement of any crop lies in exploring and harnessing the genetic diversity available in its genetic resources in the form of cultivated varieties, landraces, wild relatives, and related genera. A huge collection of wheat genetic resources is conserved in various gene banks across the globe. Molecular and phenotypic characterization followed by documentation of conserved genetic resources is a prerequisite for germplasm utilization in crop improvement. The National Genebank of India has an extensive and diverse collection of wheat germplasm, comprising Indian wheat landraces, primitive cultivars, breeding lines, and collection from other countries. The conserved germplasm can contribute immensely to the development of wheat cultivars with high levels of biotic and abiotic stress tolerance. Breeding wheat varieties that can give high yields under different stress environments has not made much headway due to high genotypes and environmental interaction, non-availability of truly resistant/tolerant germplasm, and non-availability of reliable markers linked with the QTL having a significant impact on resistance/tolerance. The development of new breeding technologies like genomic selection (GS), which takes into account the G × E interaction, will facilitate crop improvement through enhanced climate resilience, by combining biotic and abiotic stress resistance/tolerance and maximizing yield potential. In this review article, we have summarized different constraints being faced by Indian wheat-breeding programs, challenges in addressing biotic and abiotic stresses, and improving quality and nutrition. Efforts have been made to highlight the wealth of Indian wheat genetic resources available in our National Genebank and their evaluation for the identification of trait-specific germplasm. Promising genotypes to develop varieties of important targeted traits and the development of different genomics resources have also been highlighted.

Cereal grain 3D point cloud analysis method for shape extraction and filled/unfilled grain identification based on structured light imaging

Cereals are the main food for mankind. The grain shape extraction and filled/unfilled grain recognition are meaningful for crop breeding and genetic analysis. The conventional measuring method is mainly manual, which is inefficient, labor-intensive and subjective. Therefore, a novel method was proposed to extract the phenotypic traits of cereal grains based on point clouds. First, a structured light scanner was used to obtain the grains point cloud data. Then, the single grain segmentation was accomplished by image preprocessing, plane fitting, region growth clustering. The length, width, thickness, surface area and volume was calculated by the specified analysis algorithms for grain point cloud. To demonstrate this method, experimental materials included rice, wheat and corn were tested. Compared with manual measurement results, the average measurement error of grain length, width and thickness was 2.07%, 0.97%, 1.13%, and the average measurement efficiency was about 9.6 s per grain. In addition, the grain identification model was conducted with 25 grain phenotypic traits, using 6 machine learning methods. The results showed that the best accuracy for filled/unfilled grain classification was 90.184%.The best accuracy for indica and japonica identification was 99.950%, while for different varieties identification was only 47.252%. Therefore, this method was proved to be an efficient and effective way for crop research.

Agro-Morphological Characterization of Lentil Germplasm of Indian National Genebank and Development of a Core Set for Efficient Utilization in Lentil Improvement Programs

Lentil (Lens culinaris Medik.) is one of the major cool-season pulse crops worldwide. Its increasing demand as a staple pulse has led to the unlocking of diverse germplasm collections conserved in the genebanks to develop its superior varieties. The Indian National Genebank, housed at the Indian Council of Agricultural Research (ICAR)-National Bureau of Plant Genetic Resources, New Delhi, India, currently has 2,324 accessions comprising 1,796 indigenous and 528 exotic collections. This study was conducted to unveil the potential of lentil germplasm by assessing its agro-morphological characteristics and diversity, identifying trait-specific germplasm, and developing a core set. The complete germplasm set was characterized for two years, i.e., 2017-2018 and 2018-2019, and data were recorded on 26 agro-morphological traits. High phenotypic variability was observed for nine quantitative and 17 qualitative traits. A core set comprising 170 accessions (137 Indian and 33 exotic) was derived based on the characterization data as well as geographical origin using a heuristic method and PowerCore software. This core set was found to be sufficiently diverse and representative of the entire collection based on the comparison made using Shannon-Weaver diversity indices and χ2 test. These results were further validated by summary statistics. The core set displayed high genetic diversity as evident from a higher coefficient of variance in comparison to the entire set for individual traits and overall Shannon-Weaver diversity indices (entire: 1.054; core: 1.361). In addition, the total variation explained by the first three principal components was higher in the core set (70.69%) than in the entire collection (68.03%). Further, the conservation of pairwise correlation values among descriptors in the entire and core set reflected the maintenance of the structure of the whole set. Based on the results, this core set is believed to represent the entire collection, completely. Therefore, it constitutes a potential set of germplasm that can be used in the genetic enhancement of lentils.

Diversity, population structure, and linkage disequilibrium among cowpea accessions

Cowpea [Vigna unguiculata (L.) Walp] is a globally important food security crop. However, it is susceptible to pest and disease; hence, constant breeding efforts based on its diversity are required for its improvement. The present study aims to investigate the genetic diversity, population structure, and linkage disequilibrium (LD) among 274 cowpea accessions from different origins. A total of 3,127 single nucleotide polymorphism (SNP) markers generated using diversity array technology (DArT) was used. Population structure, neighbor-joining clustering, and principal component analyses indicated three subpopulations within the germplasm. Results of STRUCTURE analysis and discriminant analysis of principal components (DAPC) were complementary in assessing the structuration of the diversity among the germplasm, with the grouping of the accessions improved in DAPC. Genetic distances of 0.005-0.44 were observed among accessions. Accessions from western and central Africa, eastern and central Africa, and Asia were predominant and distributed across all subpopulations. The subpopulations had fixation indexes of 0.48-0.56. Analysis of molecular variance revealed that within subpopulation variation accounted for 81% of observed genetic variation in the germplasm. The subpopulations mainly consisted of inbred lines (inbreeding coefficient = 1) with common alleles, although they were from different geographical regions. This reflects considerable seed movement and germplasm exchange between regions. The LD was characterized by low decay for great physical distances between markers. The LD decay distance varied among chromosomes with the average distance of 80-100 kb across the genome. Thus, crop improvement is possible, and the LD will facilitate genome-wide association studies on quality attributes and critical agronomic traits in cowpea.

A review of deep learning applications for genomic selection

BACKGROUND

Several conventional genomic Bayesian (or no Bayesian) prediction methods have been proposed including the standard additive genetic effect model for which the variance components are estimated with mixed model equations. In recent years, deep learning (DL) methods have been considered in the context of genomic prediction. The DL methods are nonparametric models providing flexibility to adapt to complicated associations between data and output with the ability to adapt to very complex patterns.

MAIN BODY

We review the applications of deep learning (DL) methods in genomic selection (GS) to obtain a meta-picture of GS performance and highlight how these tools can help solve challenging plant breeding problems. We also provide general guidance for the effective use of DL methods including the fundamentals of DL and the requirements for its appropriate use. We discuss the pros and cons of this technique compared to traditional genomic prediction approaches as well as the current trends in DL applications.

CONCLUSIONS

The main requirement for using DL is the quality and sufficiently large training data. Although, based on current literature GS in plant and animal breeding we did not find clear superiority of DL in terms of prediction power compared to conventional genome based prediction models. Nevertheless, there are clear evidences that DL algorithms capture nonlinear patterns more efficiently than conventional genome based. Deep learning algorithms are able to integrate data from different sources as is usually needed in GS assisted breeding and it shows the ability for improving prediction accuracy for large plant breeding data. It is important to apply DL to large training-testing data sets.