Accuracy of genomic selection in biparental populations of flax ( Linum usitatissimum L.)

Genomic prediction for agronomic traits in a diverse Flax (Linum usitatissimum L.) germplasm collection

Breeding programs require exhaustive phenotyping of germplasms, which is time-demanding and expensive. Genomic prediction helps breeders harness the diversity of any collection to bypass phenotyping. Here, we examined the genomic prediction's potential for seed yield and nine agronomic traits using 26,171 single nucleotide polymorphism (SNP) markers in a set of 337 flax (Linum usitatissimum L.) germplasm, phenotyped in five environments. We evaluated 14 prediction models and several factors affecting predictive ability based on cross-validation schemes. Models yielded significant variation among predictive ability values across traits for the whole marker set. The ridge regression (RR) model covering additive gene action yielded better predictive ability for most of the traits, whereas it was higher for low heritable traits by models capturing epistatic gene action. Marker subsets based on linkage disequilibrium decay distance gave significantly higher predictive abilities to the whole marker set, but for randomly selected markers, it reached a plateau above 3000 markers. Markers having significant association with traits improved predictive abilities compared to the whole marker set when marker selection was made on the whole population instead of the training set indicating a clear overfitting. The correction for population structure did not increase predictive abilities compared to the whole collection. However, stratified sampling by picking representative genotypes from each cluster improved predictive abilities. The indirect predictive ability for a trait was proportionate to its correlation with other traits. These results will help breeders to select the best models, optimum marker set, and suitable genotype set to perform an indirect selection for quantitative traits in this diverse flax germplasm collection.

Sparse bayesian learning for genomic selection in yeast

Genomic selection, which predicts phenotypes such as yield and drought resistance in crops from high-density markers positioned throughout the genome of the varieties, is moving towards machine learning techniques to make predictions on complex traits that are controlled by several genes. In this paper, we consider sparse Bayesian learning and ensemble learning as a technique for genomic selection and ranking markers based on their relevance to a trait. We define and explore two different forms of the sparse Bayesian learning for predicting phenotypes and identifying the most influential markers of a trait, respectively. We apply our methods on a Saccharomyces cerevisiae dataset, and analyse our results with respect to existing related works, trait heritability, as well as the accuracies obtained from linear and Gaussian kernel functions. We find that sparse Bayesian methods are not only competitive with other machine learning methods in predicting yeast growth in different environments, but are also capable of identifying the most important markers, including both positive and negative effects on the growth, from which biologists can get insight. This attribute can make our proposed ensemble of sparse Bayesian learners favourable in ranking markers based on their relevance to a trait.

Integrated omics approaches for flax improvement under abiotic and biotic stress: Current status and future prospects

Flax (Linum usitatissimum L.) or linseed is one of the important industrial crops grown all over the world for seed oil and fiber. Besides oil and fiber, flax offers a wide range of nutritional and therapeutic applications as a feed and food source owing to high amount of α-linolenic acid (omega-3 fatty acid), lignans, protein, minerals, and vitamins. Periodic losses caused by unpredictable environmental stresses such as drought, heat, salinity-alkalinity, and diseases pose a threat to meet the rising market demand. Furthermore, these abiotic and biotic stressors have a negative impact on biological diversity and quality of oil/fiber. Therefore, understanding the interaction of genetic and environmental factors in stress tolerance mechanism and identification of underlying genes for economically important traits is critical for flax improvement and sustainability. In recent technological era, numerous omics techniques such as genomics, transcriptomics, metabolomics, proteomics, phenomics, and ionomics have evolved. The advancements in sequencing technologies accelerated development of genomic resources which facilitated finer genetic mapping, quantitative trait loci (QTL) mapping, genome-wide association studies (GWAS), and genomic selection in major cereal and oilseed crops including flax. Extensive studies in the area of genomics and transcriptomics have been conducted post flax genome sequencing. Interestingly, research has been focused more for abiotic stresses tolerance compared to disease resistance in flax through transcriptomics, while the other areas of omics such as metabolomics, proteomics, ionomics, and phenomics are in the initial stages in flax and several key questions remain unanswered. Little has been explored in the integration of omic-scale data to explain complex genetic, physiological and biochemical basis of stress tolerance in flax. In this review, the current status of various omics approaches for elucidation of molecular pathways underlying abiotic and biotic stress tolerance in flax have been presented and the importance of integrated omics technologies in future research and breeding have been emphasized to ensure sustainable yield in challenging environments.

Insights into the Genetic Architecture and Genomic Prediction of Powdery Mildew Resistance in Flax (Linum usitatissimum L.)

Powdery mildew (PM), caused by the fungus Oidium lini in flax, can cause defoliation and reduce seed yield and quality. To date, one major dominant gene (Pm1) and three quantitative trait loci (QTL) on chromosomes 1, 7 and 9 have been reported for PM resistance. To fully dissect the genetic architecture of PM resistance and identify QTL, a diverse flax core collection of 372 accessions augmented with an additional 75 breeding lines were sequenced, and PM resistance was evaluated in the field for eight years (2010–2017) in Morden, Manitoba, Canada. Genome-wide association studies (GWAS) were performed using two single-locus and seven multi-locus statistical models with 247,160 single nucleotide polymorphisms (SNPs) and the phenotypes of the 447 individuals for each year separately as well as the means over years. A total of 349 quantitative trait nucleotides (QTNs) were identified, of which 44 large-effect QTNs (R² = 10–30%) were highly stable over years. The total number of favourable alleles per accession was significantly correlated with PM resistance (r = 0.74), and genomic selection (GS) models using all identified QTNs generated significantly higher predictive ability (r = 0.93) than those constructed using the 247,160 genome-wide random SNP (r = 0.69), validating the overall reliability of the QTNs and showing the additivity of PM resistance in flax. The QTNs were clustered on the distal ends of all 15 chromosomes, especially on chromosome 5 (0.4–5.6 Mb and 9.4–16.9 Mb) and 13 (4.7–5.2 Mb). To identify candidate genes, a dataset of 3230 SNPs located in resistance gene analogues (RGAs) was used as input for GWAS, from which an additional 39 RGA-specific QTNs were identified. Overall, 269 QTN loci harboured 445 RGAs within the 200 Kb regions spanning the QTNs, including 45 QTNs located within the RGAs. These RGAs supported by significant QTN/SNP allele effects were mostly nucleotide binding site and leucine-rich repeat receptors (NLRs) belonging to either coiled-coil (CC) NLR (CNL) or toll interleukin-1 (TIR) NLR (TNL), receptor-like kinase (RLK), receptor-like protein kinase (RLP), transmembrane-coiled-coil (TM-CC), WRKY, and mildew locus O (MLO) genes. These results constitute an important genomic tool for resistance breeding and gene cloning for PM in flax.

High-throughput sequencing techniques to flax genetics and breeding

Flax (Linum usitatissimum L.) is an important oil and fiber crop. Using modern methods for flax breeding allows accelerating the introduction of some desired genes into the genotypes of future varieties. Today, an important condition for their creation is the development of research, that is based on next-generation sequencing (NGS). This review summarizes the results obtained using NGS in flax research. To date, a linkage map with a high marker density has been obtained for L. usitatissimum, which is already being used for a more efficient search for quantitative traits loci. Comparative studies of transcriptomes and miRNomes of flax under stress and in control conditions elucidated molecular-genetic mechanisms of abiotic and biotic stress responses. The very accurate model for genomic selection of flax resistant to pasmo was constructed. Based on NGS-sequencing also some details of the genus Linum evolution were clarified. The knowledge systematized in the review can be useful for researchers working in flax breeding and whereas fundamental interest for understanding the phylogenetic relationships within the genus Linum, the ontogenesis, and the mechanisms of the response of flax plants to various stress factors.

Genomic Prediction Accuracy of Seven Breeding Selection Traits Improved by QTL Identification in Flax

Molecular markers are one of the major factors affecting genomic prediction accuracy and the cost of genomic selection (GS). Previous studies have indicated that the use of quantitative trait loci (QTL) as markers in GS significantly increases prediction accuracy compared with genome-wide random single nucleotide polymorphism (SNP) markers. To optimize the selection of QTL markers in GS, a set of 260 lines from bi-parental populations with 17,277 genome-wide SNPs were used to evaluate the prediction accuracy for seed yield (YLD), days to maturity (DTM), iodine value (IOD), protein (PRO), oil (OIL), linoleic acid (LIO), and linolenic acid (LIN) contents. These seven traits were phenotyped over four years at two locations. Identification of quantitative trait nucleotides (QTNs) for the seven traits was performed using three types of statistical models for genome-wide association study: two SNP-based single-locus (SS), seven SNP-based multi-locus (SM), and one haplotype-block-based multi-locus (BM) models. The identified QTNs were then grouped into QTL based on haplotype blocks. For all seven traits, 133, 355, and 1208 unique QTL were identified by SS, SM, and BM, respectively. A total of 1420 unique QTL were obtained by SS+SM+BM, ranging from 254 (OIL, LIO) to 361 (YLD) for individual traits, whereas a total of 427 unique QTL were achieved by SS+SM, ranging from 56 (YLD) to 128 (LIO). SS models alone did not identify sufficient QTL for GS. The highest prediction accuracies were obtained using single-trait QTL identified by SS+SM+BM for OIL (0.929 ± 0.016), PRO (0.893 ± 0.023), YLD (0.892 ± 0.030), and DTM (0.730 ± 0.062), and by SS+SM for LIN (0.837 ± 0.053), LIO (0.835 ± 0.049), and IOD (0.835 ± 0.041). In terms of the number of QTL markers and prediction accuracy, SS+SM outperformed other models or combinations thereof. The use of all SNPs or QTL of all seven traits significantly reduced the prediction accuracy of traits. The results further validated that QTL outperformed high-density genome-wide random markers, and demonstrated that the combined use of single and multi-locus models can effectively identify a comprehensive set of QTL that improve prediction accuracy, but further studies on detection and removal of redundant or false-positive QTL to maximize prediction accuracy and minimize the number of QTL markers in GS are warranted.

Feature Selection Method Based on Partial Least Squares and Analysis of Traditional Chinese Medicine Data

The partial least squares method has many advantages in multivariable linear regression, but it does not include the function of feature selection. This method cannot screen for the best feature subset (referred to in this study as the "Gold Standard") or optimize the model, although contrarily using the L1 norm can achieve the sparse representation of parameters, leading to feature selection. In this study, a feature selection method based on partial least squares is proposed. In the new method, exploiting partial least squares allows extraction of the latent variables required for performing multivariable linear regression, and this method applies the L1 regular term constraint to the sum of the absolute values of the regression coefficients. This technique is then combined with the coordinate descent method to perform multiple iterations to select a better feature subset. Analyzing traditional Chinese medicine data and University of California, Irvine (UCI), datasets with the model, the experimental results show that the feature selection method based on partial least squares exhibits preferable adaptability for traditional Chinese medicine data and UCI datasets.

Evaluation of Genomic Prediction for Pasmo Resistance in Flax

Pasmo (Septoria linicola) is a fungal disease causing major losses in seed yield and quality and stem fibre quality in flax. Pasmo resistance (PR) is quantitative and has low heritability. To improve PR breeding efficiency, the accuracy of genomic prediction (GP) was evaluated using a diverse worldwide core collection of 370 accessions. Four marker sets, including three defined by 500, 134 and 67 previously identified quantitative trait loci (QTL) and one of 52,347 PR-correlated genome-wide single nucleotide polymorphisms, were used to build ridge regression best linear unbiased prediction (RR-BLUP) models using pasmo severity (PS) data collected from field experiments performed during five consecutive years. With five-fold random cross-validation, GP accuracy as high as 0.92 was obtained from the models using the 500 QTL when the average PS was used as the training dataset. GP accuracy increased with training population size, reaching values >0.9 with training population size greater than 185. Linear regression of the observed PS with the number of positive-effect QTL in accessions provided an alternative GP approach with an accuracy of 0.86. The results demonstrate the GP models based on marker information from all identified QTL and the 5-year PS average is highly effective for PR prediction.

Genome-Wide Association Study and Selection Signatures Detect Genomic Regions Associated with Seed Yield and Oil Quality in Flax

A genome-wide association study (GWAS) was performed on a set of 260 lines which belong to three different bi-parental flax mapping populations. These lines were sequenced to an averaged genome coverage of 19× using the Illumina Hi-Seq platform. Phenotypic data for 11 seed yield and oil quality traits were collected in eight year/location environments. A total of 17,288 single nucleotide polymorphisms were identified, which explained more than 80% of the phenotypic variation for days to maturity (DTM), iodine value (IOD), palmitic (PAL), stearic, linoleic (LIO) and linolenic (LIN) acid contents. Twenty-three unique genomic regions associated with 33 quantitative trait loci (QTL) for the studied traits were detected, thereby validating four genomic regions previously identified. The 33 QTL explained 48–73% of the phenotypic variation for oil content, IOD, PAL, LIO and LIN but only 8–14% for plant height, DTM and seed yield. A genome-wide selective sweep scan for selection signatures detected 114 genomic regions that accounted for 7.82% of the flax pseudomolecule and overlapped with the 11 GWAS-detected genomic regions associated with 18 QTL for 11 traits. The results demonstrate the utility of GWAS combined with selection signatures for dissection of the genetic structure of traits and for pinpointing genomic regions for breeding improvement.

Predictive ability of genomic selection models in a multi-population perennial ryegrass training set using genotyping-by-sequencing

KEY MESSAGE

Genomic prediction models for multi-year dry matter yield, via genotyping-by-sequencing in a composite training set, demonstrate potential for genetic gain improvement through within-half sibling family selection. Perennial ryegrass (Lolium perenne L.) is a key source of nutrition for ruminant livestock in temperate environments worldwide. Higher seasonal and annual yield of herbage dry matter (DMY) is a principal breeding objective but the historical realised rate of genetic gain for DMY is modest. Genomic selection was investigated as a tool to enhance the rate of genetic gain. Genotyping-by-sequencing (GBS) was undertaken in a multi-population (MP) training set of five populations, phenotyped as half-sibling (HS) families in five environments over 2 years for mean herbage accumulation (HA), a measure of DMY potential. GBS using the ApeKI enzyme yielded 1.02 million single-nucleotide polymorphism (SNP) markers from a training set of n = 517. MP-based genomic prediction models for HA were effective in all five populations, cross-validation-predictive ability (PA) ranging from 0.07 to 0.43, by trait and target population, and 0.40-0.52 for days-to-heading. Best linear unbiased predictor (BLUP)-based prediction methods, including GBLUP with either a standard or a recently developed (KGD) relatedness estimation, were marginally superior or equal to ridge regression and random forest computational approaches. PA was principally an outcome of SNP modelling genetic relationships between training and validation sets, which may limit application for long-term genomic selection, due to PA decay. However, simulation using data from the training experiment indicated a twofold increase in genetic gain for HA, when applying a prediction model with moderate PA in a single selection cycle, by combining among-HS family selection, based on phenotype, with within-HS family selection using genomic prediction.

Genetic Variability of 27 Traits in a Core Collection of Flax (Linum usitatissimum L.)

Assessment of genetic variability of plant core germplasm is needed for efficient germplasm utilization in breeding improvement. A total of 391 accessions of a flax core collection, which preserves the variation present in the world collection of 3,378 accessions maintained by Plant Gene Resources of Canada (PGRC) and represents a broad range of geographical origins, different improvement statuses and two morphotypes, was evaluated in field trials in up to 8 year-location environments for 10 agronomic, eight seed quality, six fiber and three disease resistance traits. The large phenotypic variation in this subset was explained by morphotypes (22%), geographical origins (11%), and other variance components (67%). Both divergence and similarity between two basic morphotypes, namely oil or linseed and fiber types, were observed, whereby linseed accessions had greater thousand seed weight, seeds m^-2, oil content, branching capability and resistance to powdery mildew while fiber accessions had greater straw weight, plant height, protein content and resistance to pasmo and fusarium wilt diseases, but they had similar performance in many traits and some of them shared common characteristics of fiber and linseed types. Weak geographical patterns within either fiber or linseed accessions were confirmed, but specific trait performance was identified in East Asia for fiber type, and South Asia and North America for linseed type. Relatively high broad-sense heritability was obtained for seed quality traits, followed by agronomic traits and resistance to powdery mildew and fusarium wilt. Diverse phenotypic and genetic variability in the flax core collection constitutes a useful resource for breeding.