Improvement of genomic prediction in advanced wheat breeding lines by including additive-by-additive epistasis

Key message Including additive and additive-by-additive epistasis in a NOIA parametrization did not yield orthogonal partitioning of genetic variances, nevertheless, it improved predictive ability in a leave-one-out cross-validation for wheat grain yield. Abstract Additive-by-additive epistasis is the principal non-additive genetic effect in inbred wheat lines and is potentially useful for developing cultivars based on total genetic merit; nevertheless, its practical benefits have been highly debated. In this article, we aimed to (i) evaluate the performance of models including additive and additive-by-additive epistatic effects for variance components (VC) estimation of grain yield in a wheat-breeding population, and (ii) to investigate whether including additive-by-additive epistasis in genomic prediction enhance wheat grain yield predictive ability (PA). In total, 2060 sixth-generation (F6) lines from Nordic Seed A/S breeding company were phenotyped in 21 year-location combinations in Denmark, and genotyped using a 15 K-Illumina-BeadChip. Three models were used to estimate VC and heritability at plot level: (i) “I-model” (baseline), (ii) “I + GA-model”, extending I-model with an additive genomic effect, and (iii) “I + GA + GAA-model”, extending I + GA-model with an additive-by-additive genomic effects. The I + GA-model and I + GA + GAA-model were based on the Natural and Orthogonal Interactions Approach (NOIA) parametrization. The I + GA + GAA-model failed to achieve orthogonal partition of genetic variances, as revealed by a change in estimated additive variance of I + GA-model when epistasis was included in the I + GA + GAA-model. The PA was studied using leave-one-line-out and leave-one-breeding-cycle-out cross-validations. The I + GA + GAA-model increased PA significantly (16.5%) compared to the I + GA-model in leave-one-line-out cross-validation. However, the improvement due to including epistasis was not observed in leave-one-breeding-cycle-out cross-validation. We conclude that epistatic models can be useful to enhance predictions of total genetic merit. However, even though we used the NOIA parameterization, the variance partition into orthogonal genetic effects was not possible.

Significance of Linkage Disequilibrium and Epistasis on the Genetic Variances in Non- Inbred and Inbred Populations

Background The influence of linkage disequilibrium (LD), epistasis, and inbreeding on the genotypic variance continues to be an important area of investigation in genetics and evolution. Although the current knowledge about biological pathways and gene networks imply that epistasis is important in determining quantitative traits, the empirical evidence for a range of species and traits is that the genetic variance is most additive. This is confirmed by some recent theoretical studies. However, because these investigations have assumed linkage equilibrium, only additive effects, or simplified assumptions for the two- and high-order epistatic effects, the objective of this investigation was to provide additional information about the impact of LD and epistasis on the genetic variances in non-inbred and inbred populations, using a simulated data set.Results The epistatic variance in generation 0 corresponded to 1 to 10% of the genotypic variance, with 30% of epistatic genes, but it corresponded to 5 to 45% assuming 100% of epistatic genes. After 10 generations of random cross or selfing the ratio epistatic variance/genotypic variance increased in the range of 15 to 1,079%. The epistatic variances are maximized assuming dominant epistasis, duplicate genes with cumulative effects, and non-epistatic gene interaction. A minimization occurs with complementary, recessive, and dominant and recessive epistasis. In non-inbred populations, the genetic covariances have negligible magnitude compared with the genetic variances. In inbred populations, excepting for duplicate epistasis, the sum of the epistatic covariances was in general negative and with magnitude higher than the non-additive variances, especially under 100% of epistatic genes.Conclusions The LD level for genes, even under a relatively low gene density, has a significant effect on the genetic variances in non-inbred and inbred populations. Assuming digenic epistasis, the additive variance is in general the most important component of the genotypic variance in non-inbred and inbred populations. The ratio epistatic variance/genotypic variance is proportional to the percentage of interacting genes and increases with random cross and selfing. In general, the additive x additive variance is the most important component of the epistatic variance. The maximization of the epistatic variance depends on the allele frequency, LD level, and epistasis type.

Significance of Linkage Disequilibrium and Epistasis on the Genetic Variances in Non-Inbred and Inbred Populations

Background The influence of linkage disequilibrium (LD), epistasis, and inbreeding on the genotypic variance continues to be an important area of investigation in genetics and evolution. Although the current knowledge about biological pathways and gene networks imply that epistasis is important in determining quantitative traits, the empirical evidence for a range of species and traits is that the genetic variance is most additive. This is confirmed by some recent theoretical studies. However, because these investigations have assumed linkage equilibrium, only additive effects, or simplified assumptions for the two- and high-order epistatic effects, the objective of this investigation was to provide additional information about the impact of LD and epistasis on the genetic variances in non-inbred and inbred populations, using a simulated data set.Results The epistatic variance in generation 0 corresponded to 1 to 10% of the genotypic variance, with 30% of epistatic genes, but it corresponded to 5 to 45% assuming 100% of epistatic genes. After 10 generations of random cross or selfing the ratio epistatic variance/genotypic variance increased in the range of 15 to 1,079%. The epistatic variances are maximized assuming dominant epistasis, duplicate genes with cumulative effects, and non-epistatic gene interaction. A minimization occurs with complementary, recessive, and dominant and recessive epistasis. In non-inbred populations, the genetic covariances have negligible magnitude compared with the genetic variances. In inbred populations, excepting for duplicate epistasis, the sum of the epistatic covariances was in general negative and with magnitude higher than the non-additive variances, especially under 100% of epistatic genes.Conclusions The LD level for genes, even under a relatively low gene density, has a significant effect on the genetic variances in non-inbred and inbred populations. Assuming digenic epistasis, the additive variance is in general the most important component of the genotypic variance in non-inbred and inbred populations. The ratio epistatic variance/genotypic variance is proportional to the percentage of interacting genes and increases with random cross and selfing. In general, the additive x additive variance is the most important component of the epistatic variance. The maximization of the epistatic variance depends on the allele frequency, LD level, and epistasis type.

130 Utilization of Nutrack to Assess Variance Components and Heritability of Activity Traits

Overall activity and behavior are integral components of sows remaining productive in the herd. This investigation studied overall activity of group housed replacement gilts and the heritability of various activity traits. Beginning around 20 wk of age, video recorded data of approximately 75 gilts/group for a total of 2,378 gilts over 32 groups was collected for 7 consecutive d using the NUtrack System, which tracks distance travelled (m), avg speed (m/s), angle rotated (degrees), and time standing (s), sitting (s), eating (s), and laying (s). The recorded phenotypes were standardized to the distribution observed within a pen for each group. The final values used for analysis were the average daily standardized values. Data were analyzed using mixed models (RStudio V 1.2.5033) including effects of sire, dam, dam’s sire and dam, dam’s grandsire and granddam, farrowing group, barn, pen, and on-test date. Sire had an effect on every activity trait P < 0.001), and dam had an effect on average speed (P < 0.001). The dam’s sire had an effect on all activity traits (P < 0.001) and the dam’s grandsire had an effect on average speed (P < 0.001). Heritabilities and variance components of activity traits were estimated in ASReml 4 using an animal model with a two-generation pedigree. Genetic variances are 0.17 +/- 0.029, 0.19 +/- 0.034, and 0.11 +/- 0.024, residual variances are 0.37 +/- 0.023, 0.41 +/- 0.027, and 0.41 +/- 0.022, phenotypic variances are 0.54 +/- 0.018, 0.60 +/- 0.020, and 0.52 +/- 0.016, and heritabilities are 0.32 +/- 0.048, 0.32 +/- 0.049, and 0.21 +/- 0.044 for average speed, distance, and lie respectively. NUtrack offers potential to aid in selection decisions. Given the results presented herein, continued investigation into these activity traits and their association with sow longevity is warranted.

The predictive potential of key adaptation parameters and proxy fitness traits between benign and stressful thermal environments

Understanding the adaptive potential of a species is key when predicting whether a species can contend with climate change. Adaptive capacity depends on the amount of genetic variation within a population for relevant traits. However, genetic variation changes in different environments, making it difficult to predict whether a trait will respond to selection when not measured directly in that environment. Here, we investigated how genetic variances, and phenotypic and genetic covariances, between a fitness trait and morphological traits changed between thermal environments in two closely-related Drosophila. If morphological traits strongly correlate with fitness, they may provide an easy-to-measure proxy of fitness to aid in understanding adaptation potential. We used a parent-offspring quantitative genetic design to test the effect of a benign (23°C) and stressful (28°C) thermal environment on genetic variances of fecundity and wing size and shape, and their phenotypic and genetic covariances. We found genetic variances were higher within the stressful environment for fecundity but lower within the stressful environment for wing size. We did not find evidence for significant phenotypic correlations. Phenotypic and genetic correlations did not reveal a consistent pattern between thermal environments or within or between species. This corroborates previous research and reiterates that conclusions drawn in one environment about the adaptive potential of a trait, and the relationship of that trait with fitness, cannot be extrapolated to other environments or within or between closely-related species. This confirms that researchers should use caution when generalising findings across environments in terms of genetic variation and adaptive potential.

Significance of linkage disequilibrium and epistasis on the genetic variances and covariance between relatives in non-inbred and inbred populations

Because no feasible theoretical model can depict the complexity of phenotype development from a genotype, the joint significance of linkage disequilibrium (LD), epistasis, and inbreeding on the genetic variances remains unclear. The objective of this investigation was to assess the impact of LD and epistasis on the genetic variances and covariances between relatives in non-inbred and inbred populations using simulated data. We provided the theoretical background and simulated grain yield assuming 400 genes in 10 chromosomes of 200 and 50 cM. We generated five populations with low to high LD levels, assuming 10 generations of random cross and selfing. The analysis of the parametric LD in the populations shows that the LD level depends mainly on the gene density. The significance of the LD level is impressive on the magnitude of the genotypic and additive variances, which is the most important component of the genotypic variance, regardless of the LD level and the degree of inbreeding. Regardless of the type of epistasis, the ratio epistatic variance/genotypic variance is proportional to the percentage of the epistatic genes. For the epistatic variances, except for duplicate epistasis and dominant and recessive epistasis, with 100% of epistatic genes, their magnitudes are much lower than the magnitude of the additive variance. The additive x additive variance is the most important epistatic variance. Our results explain why LD for genes and relationship information are key factors affecting the genomic prediction accuracy of complex traits and the efficacy of association studies.

Genomic mating in outbred species: predicting cross usefulness with additive and total genetic covariance matrices

Diverse crops are both outbred and clonally propagated. Breeders typically use truncation selection of parents and invest significant time, land and money evaluating the progeny of crosses to find exceptional genotypes. We developed and tested genomic mate selection criteria suitable for organisms of arbitrary homozygosity level where the full-sibling progeny are of direct interest as future parents and/or cultivars. We extended cross variance and covariance variance prediction to include dominance effects and predicted the multivariate selection index genetic variance of crosses based on haplotypes of proposed parents, marker effects and recombination frequencies. We combined the predicted mean and variance into usefulness criteria for parent and variety development. We present an empirical study of cassava (Manihot esculenta), a staple tropical root crop. We assessed the potential to predict the multivariate genetic distribution (means, variances and trait covariances) of 462 cassava families in terms of additive and total value using cross-validation. We were able to predict all genetic variances and most covariances with non-zero accuracy. We also tested a directional dominance model and found significant inbreeding depression for most traits and a boost in total merit accuracy for root yield. We predicted 47,083 possible crosses of 306 parents and contrasted them to those previously tested to show how mate selection can reveal new potential within the germplasm. We enable breeders to consider the potential of crosses to produce future parents (progeny with excellent breeding values) and varieties (progeny with top performance).Author SummaryBreeders typically use truncation selection and invest significant resources evaluating progeny to find exceptional genotypes. We extended genetic variance and trait covariance prediction to include dominance and predicting the multivariate selection index variance. We enable mate selection based on potential to produce future parents (progeny with excellent breeding values) and/or varieties (progeny with top performance). Using cross-validation, we demonstrate that genetic variances and covariances can be predicted with non-zero accuracy in cassava, a staple tropical root crop.

Changes of genetic variances and heritability by effect of selection in a Mexican local variety of Squash

A local variety of squash (Cucurbita pepo L.), ‘Round Zuchinni’ type from Los Reyes Acatlixhuayan State of México, México, was simultaneously evaluated and selected. The objective was to study effect of selection on genetic variance and heritability. Additive genetic variance decreased in seven of nine traits studied: fruit height (79.93 %); fruit weight (65.72 %); fruit width (60.91 %); flesh thickness (57.66 %); flesh color (43.70 %); dry weight of seed (39.54 %); flesh flavor (16.60 %); except in width and seed length traits where it increased 63.40 % and 0.81 %, respectively. Only weight of seed had dominance genetic variance. The coefficient of additive genetic variance (CVA) fluctuated from 9.4 to 61.7 % in the first cycle, and from 9.4 to 51.8 % in the second cycle of selection-evaluation among traits. Heritability diminished in seven from nine characters. In general, the estimated genetic variances (additive and dominance) and heritability decreased as a result of combined selection of falf sib families.

Accurate Estimation of Marker-Associated Genetic Variance and Heritability in Complex Trait Analyses

ABSTRACTThe emergence of high-throughput, genome-scale approaches for identifying and genotyping DNA variants has been a catalyst for the development of increasingly sophisticated whole-genome association and genomic prediction approaches, which together have revolutionized the study of complex traits in human, animal, and plant populations. These approaches have uncovered a broad spectrum of genetic complexity across traits and organisms, from a small number of detectable loci to an unknown number of undetectable loci. The heritable variation observed in a population is often partly caused by the segregation of one or more large-effect (statistically detectable) loci. Our study focused on the accurate estimation of the proportion of the genetic variance explained by such loci (p), a parameter estimated to quantify and predict the importance of causative loci or markers in linkage disequilibrium with causative loci. Here, we show that marker-associated genetic variances are systematically overestimated by standard statistical methods. The upward bias is purely mathematical in nature, unrelated to selection bias, and caused by the inequality between the genetic variance among progeny and sums of partitioned marker-associated genetic variances. We discovered a straightforward mathematical correction factor (kM) that depends only on degrees of freedom and the number of entries, is constant for a given experiment design, expands to higher-order genetic models in a predictable pattern, and yields bias-corrected estimates of marker-associated genetic variance and heritability.