A Bayesian Genomic Multi-output Regressor Stacking Model for Predicting Multi-trait Multi-environment Plant Breeding Data

In this paper we propose a Bayesian multi-output regressor stacking (BMORS) model that is a generalization of the multi-trait regressor stacking method. The proposed BMORS model consists of two stages: in the first stage, a univariate genomic best linear unbiased prediction (GBLUP including genotype × environment interaction GE) model is implemented for each of the L traits under study; then the predictions of all traits are included as covariates in the second stage, by implementing a Ridge regression model. The main objectives of this research were to study alternative models to the existing multi-trait multi-environment (BMTME) model with respect to (1) genomic-enabled prediction accuracy, and (2) potential advantages in terms of computing resources and implementation. We compared the predictions of the BMORS model to those of the univariate GBLUP model using 7 maize and wheat datasets. We found that the proposed BMORS produced similar predictions to the univariate GBLUP model and to the BMTME model in terms of prediction accuracy; however, the best predictions were obtained under the BMTME model. In terms of computing resources, we found that the BMORS is at least 9 times faster than the BMTME method. Based on our empirical findings, the proposed BMORS model is an alternative for predicting multi-trait and multi-environment data, which are very common in genomic-enabled prediction in plant and animal breeding programs.

Download Full-text

Accounting for Group-Specific Allele Effects and Admixture in Genomic Predictions: Theory and Experimental Evaluation in Maize

Genetics ◽

10.1534/genetics.120.303278 ◽

2020 ◽

Vol 216 (1) ◽

pp. 27-41

Author(s):

Simon Rio ◽

Laurence Moreau ◽

Alain Charcosset ◽

Tristan Mary-Huard

Keyword(s):

Genomic Prediction ◽

Prediction Accuracy ◽

Prediction Models ◽

Best Linear Unbiased Prediction ◽

Linear Unbiased Prediction ◽

Modeling Group ◽

A Genome ◽

Specific Allele ◽

Best Linear Unbiased ◽

Unbiased Prediction

Populations structured into genetic groups may display group-specific linkage disequilibrium, mutations, and/or interactions between quantitative trait loci and the genetic background. These factors lead to heterogeneous marker effects affecting the efficiency of genomic prediction, especially for admixed individuals. Such individuals have a genome that is a mosaic of chromosome blocks from different origins, and may be of interest to combine favorable group-specific characteristics. We developed two genomic prediction models adapted to the prediction of admixed individuals in presence of heterogeneous marker effects: multigroup admixed genomic best linear unbiased prediction random individual (MAGBLUP-RI), modeling the ancestry of alleles; and multigroup admixed genomic best linear unbiased prediction random allele effect (MAGBLUP-RAE), modeling group-specific distributions of allele effects. MAGBLUP-RI can estimate the segregation variance generated by admixture while MAGBLUP-RAE can disentangle the variability that is due to main allele effects from the variability that is due to group-specific deviation allele effects. Both models were evaluated for their genomic prediction accuracy using a maize panel including lines from the Dent and Flint groups, along with admixed individuals. Based on simulated traits, both models proved their efficiency to improve genomic prediction accuracy compared to standard GBLUP models. For real traits, a clear gain was observed at low marker densities whereas it became limited at high marker densities. The interest of including admixed individuals in multigroup training sets was confirmed using simulated traits, but was variable using real traits. Both MAGBLUP models and admixed individuals are of interest whenever group-specific SNP allele effects exist.

Download Full-text

Genomic studies with pre-selected markers reveal dominance effects influencing growth traits in Eucalyptus nitens

G3 Genes|Genome|Genetics ◽

10.1093/g3journal/jkab363 ◽

2021 ◽

Author(s):

Bala R Thumma ◽

Kelsey R Joyce ◽

Andrew Jacobs

Keyword(s):

Inbreeding Depression ◽

Prediction Accuracy ◽

Basic Density ◽

Predictive Ability ◽

Best Linear Unbiased Prediction ◽

The Other ◽

Eucalyptus Nitens ◽

Linear Unbiased Prediction ◽

Best Linear Unbiased ◽

Unbiased Prediction

Abstract Genomic selection (GS) is being increasingly adopted by the tree breeding community. Most of the GS studies in trees are focused on estimating additive genetic effects. Exploiting the dominance effects offers additional opportunities to improve genetic gain. To detect dominance effects, trait relevant markers may be important compared to non-selected markers. Here we used pre-selected markers to study the dominance effects in a Eucalyptus nitens (E. nitens) breeding population consisting of open-pollinated (OP) and controlled-pollinated (CP) families. We used 8221 trees from six progeny trials in this study. Of these, 868 progeny and 255 parents were genotyped with the E. nitens marker panel. Three traits; diameter at breast height (DBH), wood basic density (DEN) and kraft pulp yield (KPY) were analysed. Two types of genomic relationship matrices based on identity-by-state (IBS) and identity-by-descent (IBD) were tested. Performance of the genomic best linear unbiased prediction (GBLUP) models with IBS and IBD matrices were compared with pedigree-based additive best linear unbiased prediction (ABLUP) models with and without the pedigree reconstruction. Similarly, the performance of the single-step GBLUP (ssGBLUP) with IBS and IBD matrices were compared with ABLUP models using all 8221 trees. Significant dominance effects were observed with the GBLUP-AD model for DBH. The predictive ability of DBH is higher with the GBLUP-AD model compared to other models. Similarly, the prediction accuracy of genotypic values is higher with GBLUP-AD compared to the GBLUP-A model. Among the two GBLUP models (IBS and IBD), no differences were observed in predictive abilities and prediction accuracies. While the estimates of predictive ability with additive effects were similar among all four models, prediction accuracies of ABLUP were lower than the GBLUP models. The prediction accuracy of ssGBLUP-IBD is higher than the other three models while the theoretical accuracy of ssGBLUP-IBS is consistently higher than the other three models across all three groups tested (parents, genotyped, non-genotyped). Significant inbreeding depression was observed for DBH and KPY. While there is a linear relationship between inbreeding and DBH, the relationship between inbreeding and KPY is non-linear and quadratic. These results indicate that the inbreeding depression of DBH is mainly due to directional dominance while in KPY it may be due to epistasis. Inbreeding depression may be the main source of the observed dominance effects in DBH. The significant dominance effect observed for DBH may be used to select complementary parents to improve the genetic merit of the progeny in E. nitens.

Download Full-text

Efficient Computation of Ridge-Regression Best Linear Unbiased Prediction in Genomic Selection in Plant Breeding

Crop Science ◽

10.2135/cropsci2011.11.0592 ◽

2012 ◽

Vol 52 (3) ◽

pp. 1093-1104 ◽

Cited By ~ 26

Author(s):

H. P. Piepho ◽

J. O. Ogutu ◽

T. Schulz-Streeck ◽

B. Estaghvirou ◽

A. Gordillo ◽

...

Keyword(s):

Plant Breeding ◽

Genomic Selection ◽

Ridge Regression ◽

Best Linear Unbiased Prediction ◽

Efficient Computation ◽

Linear Unbiased Prediction ◽

Best Linear Unbiased ◽

Unbiased Prediction

Download Full-text

Core-dependent changes in genomic predictions using the Algorithm for Proven and Young in single-step genomic best linear unbiased prediction

Journal of Animal Science ◽

10.1093/jas/skaa374 ◽

2020 ◽

Vol 98 (12) ◽

Author(s):

Ignacy Misztal ◽

Shogo Tsuruta ◽

Ivan Pocrnic ◽

Daniela Lourenco

Keyword(s):

Prediction Accuracy ◽

Best Linear Unbiased Prediction ◽

Single Step ◽

Relationship Matrix ◽

Linear Unbiased Prediction ◽

Breeding Values ◽

Best Linear Unbiased ◽

The Impact ◽

Genomic Predictions ◽

Unbiased Prediction

Abstract Single-step genomic best linear unbiased prediction with the Algorithm for Proven and Young (APY) is a popular method for large-scale genomic evaluations. With the APY algorithm, animals are designated as core or noncore, and the computing resources to create the inverse of the genomic relationship matrix (GRM) are reduced by inverting only a portion of that matrix for core animals. However, using different core sets of the same size causes fluctuations in genomic estimated breeding values (GEBVs) up to one additive standard deviation without affecting prediction accuracy. About 2% of the variation in the GRM is noise. In the recursion formula for APY, the error term modeling the noise is different for every set of core animals, creating changes in breeding values. While average changes are small, and correlations between breeding values estimated with different core animals are close to 1.0, based on the normal distribution theory, outliers can be several times bigger than the average. Tests included commercial datasets from beef and dairy cattle and from pigs. Beyond a certain number of core animals, the prediction accuracy did not improve, but fluctuations decreased with more animals. Fluctuations were much smaller than the possible changes based on prediction error variance. GEBVs change over time even for animals with no new data as genomic relationships ties all the genotyped animals, causing reranking of top animals. In contrast, changes in nongenomic models without new data are small. Also, GEBV can change due to details in the model, such as redefinition of contemporary groups or unknown parent groups. In particular, increasing the fraction of blending of the GRM with a pedigree relationship matrix from 5% to 20% caused changes in GEBV up to 0.45 SD, with a correlation of GEBV > 0.99. Fluctuations in genomic predictions are part of genomic evaluation models and are also present without the APY algorithm when genomic evaluations are computed with updated data. The best approach to reduce the impact of fluctuations in genomic evaluations is to make selection decisions not on individual animals with limited individual accuracy but on groups of animals with high average accuracy.

Download Full-text

Seleção genômica para melhoramento vegetal com diferentes estruturas populacionais

Pesquisa Agropecuária Brasileira ◽

10.1590/s0100-204x2016001100008 ◽

2016 ◽

Vol 51 (11) ◽

pp. 1857-1867 ◽

Cited By ~ 2

Author(s):

Mágno Sávio Ferreira Valente ◽

◽

José Marcelo Soriano Viana ◽

Marcos Deon Vilela de Resende ◽

Fabyano Fonseca e Silva ◽

...

Keyword(s):

Ridge Regression ◽

Best Linear Unbiased Prediction ◽

Linear Unbiased Prediction ◽

Best Linear Unbiased ◽

Unbiased Prediction

Resumo O objetivo deste trabalho foi avaliar a eficiência da seleção genômica em diferentes cenários de estrutura populacional em milho-pipoca, com estimação dos efeitos e uso de marcadores na própria população de referência e em populações não relacionadas, e determinar a influência do tamanho efetivo e das relações de parentesco na população de estimação sobre a acurácia da predição. Foram simuladas populações com diferentes desequilíbrios de ligação (LD) e variâncias aditivas, tendo-se considerado diferentes caracteres, densidades de marcadores, herdabilidades e gerações, no total de 144 cenários. Também foram simuladas populações estruturadas em progênies. A acurácia da predição dos valores genéticos aditivos foi obtida por meio da correlação entre os valores paramétricos e os valores estimados por RR-BLUP (ridge regression-best linear unbiased prediction). Em populações com baixo LD e menor variância aditiva, o uso de maiores densidades de SNP (10 SNP 0,1 cM-1) é indicado, e, além disso, o candidato à seleção deve ser relacionado à população de estimação, para que a acurácia de predição seja satisfatória. O uso de população de seleção na mesma geração da população de estimação reduz em pelo menos 8% a acurácia. A estruturação da população em progênies de maior relacionamento e menor tamanho efetivo aumenta a eficiência da seleção genômica.

Download Full-text

Improving the accuracy of genomic evaluation for linear body measurement traits using single-step genomic best linear unbiased prediction in Hanwoo beef cattle

BMC Genetics ◽

10.1186/s12863-020-00928-1 ◽

2020 ◽

Vol 21 (1) ◽

Author(s):

Masoumeh Naserkheil ◽

Deuk Hwan Lee ◽

Hossein Mehrban

Keyword(s):

Prediction Accuracy ◽

Best Linear Unbiased Prediction ◽

Single Step ◽

Farm Animals ◽

Body Measurement ◽

Data Set ◽

Linear Unbiased Prediction ◽

Body Measurement Traits ◽

Best Linear Unbiased ◽

Unbiased Prediction

Abstract Background Recently, there has been a growing interest in the genetic improvement of body measurement traits in farm animals. They are widely used as predictors of performance, longevity, and production traits, and it is worthwhile to investigate the prediction accuracies of genomic selection for these traits. In genomic prediction, the single-step genomic best linear unbiased prediction (ssGBLUP) method allows the inclusion of information from genotyped and non-genotyped relatives in the analysis. Hence, we aimed to compare the prediction accuracy obtained from a pedigree-based BLUP only on genotyped animals (PBLUP-G), a traditional pedigree-based BLUP (PBLUP), a genomic BLUP (GBLUP), and a single-step genomic BLUP (ssGBLUP) method for the following 10 body measurement traits at yearling age of Hanwoo cattle: body height (BH), body length (BL), chest depth (CD), chest girth (CG), chest width (CW), hip height (HH), hip width (HW), rump length (RL), rump width (RW), and thurl width (TW). The data set comprised 13,067 phenotypic records for body measurement traits and 1523 genotyped animals with 34,460 single-nucleotide polymorphisms. The accuracy for each trait and model was estimated only for genotyped animals using five-fold cross-validations. Results The accuracies ranged from 0.02 to 0.19, 0.22 to 0.42, 0.21 to 0.44, and from 0.36 to 0.55 as assessed using the PBLUP-G, PBLUP, GBLUP, and ssGBLUP methods, respectively. The average predictive accuracies across traits were 0.13 for PBLUP-G, 0.34 for PBLUP, 0.33 for GBLUP, and 0.45 for ssGBLUP methods. Our results demonstrated that averaged across all traits, ssGBLUP outperformed PBLUP and GBLUP by 33 and 43%, respectively, in terms of prediction accuracy. Moreover, the least root of mean square error was obtained by ssGBLUP method. Conclusions Our findings suggest that considering the ssGBLUP model may be a promising way to ensure acceptable accuracy of predictions for body measurement traits, especially for improving the prediction accuracy of selection candidates in ongoing Hanwoo breeding programs.

Download Full-text

GENETIC DIVERGENCE AMONG COWPEA GENOTYPES BASED ON CHARACTERS ASSOCIATED WITH THE GREEN POD AND GRAIN MARKET

Revista Caatinga ◽

10.1590/1983-21252018v31n107rc ◽

2018 ◽

Vol 31 (1) ◽

pp. 56-63

Author(s):

JOSÉ TORRES FILHO ◽

CHRISTIANE NORONHA GOMES DOS SANTOS OLIVEIRA ◽

LINDOMAR MARIA DA SILVEIRA ◽

GLAUBER HENRIQUE DE SOUSA NUNES ◽

CARLA CAROLINE ALVES PEREIRA ◽

...

Keyword(s):

Genetic Divergence ◽

Block Design ◽

Joint Analysis ◽

Group Method ◽

Environment Interaction ◽

Linear Unbiased Prediction ◽

Pair Group ◽

Genotype Environment Interaction ◽

Best Linear Unbiased ◽

Grain Market

ABSTRACT The objective of this study was to evaluate genetic divergence among cowpea genotypes and to select parents for crosses aimed at the fresh pod and grain market. Two experiments were carried out during 2014, corresponding to two sowing times, in the municipality of Mossoró, State of Rio Grande do Norte. Twenty-three cowpea genotypes were evaluated in a randomized complete block design with four replicates. Fifteen descriptors were used to quantify divergence, using the Mahalanobis distance as a measure of dissimilarity, obtained from the genotypic mean predicted by the Restricted Maximum Likelihood/Best Linear Unbiased Prediction (REML/BLUP) method. The Unweighted Pair Group Method with Arithmetic Mean (UPGMA) hierarchical method was used to group genotypes and the Singh criterion to quantify the contribution of traits to genetic divergence. The genotype × environment interaction (G × E) influenced divergence, both in the contribution of traits and in the grouping of genotypes. In the experiment 1, the genotypes were distributed among four groups. In the experiment 2, less discrimination occurred and the genotypes were represented by only two groups. When joint analysis of two evaluations was carried out based on two sowing times, genotypes were represented by six groups. The G × E interaction influences the contribution of traits and the grouping of cowpea genotypes in the study of divergence. The genetic divergence of the set of cowpea genotypes evaluated is mainly due to green grain and pod yield. BRS Aracê and BRS Xiquexique cultivars are the most divergent among the genotypes studied, representing 75% of the recommended crosses.

Download Full-text

Genetic Variation in Drought-Tolerance Traits and Their Relationships to Growth in Pinus radiata D. Don Under Water Stress

Frontiers in Plant Science ◽

10.3389/fpls.2021.766803 ◽

2022 ◽

Vol 12 ◽

Author(s):

Ahmed Ismael ◽

Jianming Xue ◽

Dean Francis Meason ◽

Jaroslav Klápště ◽

Marta Gallart ◽

...

Keyword(s):

Genetic Variation ◽

Water Stress ◽

Genetic Correlations ◽

Best Linear Unbiased Prediction ◽

Linear Unbiased Prediction ◽

Breeding Programs ◽

Water Stress Tolerance ◽

Drought Tolerant ◽

Best Linear Unbiased ◽

Selection Of

The selection of drought-tolerant genotypes is globally recognized as an effective strategy to maintain the growth and survival of commercial tree species exposed to future drought periods. New genomic selection tools that reduce the time of progeny trials are required to substitute traditional tree breeding programs. We investigated the genetic variation of water stress tolerance in New Zealand-grown Pinus radiata D. Don using 622 commercially-used genotypes from 63 families. We used quantitative pedigree-based (Genomic Best Linear Unbiased Prediction or ABLUP) and genomic-based (Genomic Best Linear Unbiased Prediction or GBLUP) approaches to examine the heritability estimates associated with water stress tolerance in P. radiata. Tree seedling growth traits, foliar carbon isotope composition (δ13C), and dark-adapted chlorophyll fluorescence (Y) were monitored before, during and after 10 months of water stress. Height growth showed a constant and moderate heritability level, while the heritability estimate for diameter growth and δ13C decreased with water stress. In contrast, chlorophyll fluorescence exhibited low heritability after 5 and 10 months of water stress. The GBLUP approach provided less breeding value accuracy than ABLUP, however, the relative selection efficiency of GBLUP was greater compared with ABLUP selection techniques. Although there was no significant relationship directly between δ13C and Y, the genetic correlations were significant and stronger for GBLUP. The positive genetic correlations between δ13C and tree biomass traits under water stress indicated that intraspecific variation in δ13C was likely driven by differences in the genotype’s photosynthetic capacity. The results show that foliar δ13C can predict P. radiata genotype tolerance to water stress using ABLUP and GBLUP approaches and that such approaches can provide a faster screening and selection of drought-tolerant genotypes for forestry breeding programs.

Download Full-text

Genetic Parameters and Selection in Full-Sib Families of Tall Fescue Using Best Linear Unbiased Prediction (BLUP) Analysis of Animal Breeding and Genetic

10.21203/rs.3.rs-1198310/v1 ◽

2021 ◽

Author(s):

Fatemeh Pirnajmedin ◽

Mohammad Mahdi Majidi ◽

Mohammad Hadi Taleb ◽

Davoud Rostami

Keyword(s):

Ssr Markers ◽

Tall Fescue ◽

Genetic Parameters ◽

Best Linear Unbiased Prediction ◽

Forage Yield ◽

Linear Unbiased Prediction ◽

Breeding Programs ◽

Sib Families ◽

Best Linear Unbiased ◽

Selection Of

Abstract Background: Better understanding of genetic structure of economic traits is crucial for identification and selection of superior genotypes in specific breeding programs. Best linear unbiased prediction (BLUP) is the most efficient method in this regards, which is poorly used in forage plant breeding. The present study aimed to assess genetic variation, estimate genetic parameters, and predict breeding values of five essential traits in full sib families (recognized by EST-SSR markers) of tall fescue using REML/BLUP procedure. Method: Forty-two full-sib families of tall fescue (included of 120 individual genotypes), recognized by EST-SSR markers’ along with twenty-one their corresponding parental genotypes were assessed for biomass production and agro-morphological traits at three harvests (spring, summer, and autumn) in the field during 4 years (2017-2020). Results: Considerable genotypic variability was observed for all traits. Low narrow-sense heritability (h2n) for dry forage yield (DFY) at three harvest indicates that non-additive gene actions may play an important role in the inheritance of this trait. Higher h2n of yield related traits and flowering time and also significant genetic correlation of these traits with forage yield, suggests that selection based on these traits via developing an index may lead to indirect genetic improvement of DFY. Conclusion: Our results showed the adequacy of REML/BLUP procedure for identification and selection of preferable parental genotypes and progenies with higher breeding values for future breeding programs such as variety development in tall fescue. Parental genotypes 21M, 1M, and 20L were identified as superior and stable genotypes and could also produce the best hybrid combinations when they were mostly used as maternal parent.

Download Full-text

Quantitative genetic analysis of methylxanthines and phenolic compounds in mate progenies

Pesquisa Agropecuária Brasileira ◽

10.1590/s0100-204x2010000200008 ◽

2010 ◽

Vol 45 (2) ◽

pp. 171-177 ◽

Cited By ~ 8

Author(s):

Euclides Lara Cardozo Junior ◽

Carmen Maria Donaduzzi ◽

Osvaldo Ferrarese-Filho ◽

Juliana Cristhina Friedrich ◽

Adriana Gonela ◽

...

Keyword(s):

Phenolic Compounds ◽

High Performance ◽

Genetic Parameters ◽

Best Linear Unbiased Prediction ◽

Region Of Origin ◽

Linear Unbiased Prediction ◽

Breeding Programs ◽

Quantitative Genetic ◽

Sib Families ◽

Best Linear Unbiased

The objective of this work was to determine the contents of methylxanthines, caffeine and theobromine, and phenolic compounds, chlorogenic and caffeic acids, in 51 mate progenies (half-sib families) and estimate the heritability of genetic parameters. Mate progenies were from five Brazilian municipalities: Pinhão, Ivaí, Barão de Cotegipe, Quedas do Iguaçu, and Cascavel. The progenies were grown in the Ivaí locality. The contents of the compounds were obtained by high performance liquid chromatography (HPLC). The estimation of genetic parameters by the restricted maximum likelihood (REML) and the prediction of genotypic values via best linear unbiased prediction (BLUP) were obtained by the Selegen - REML/BLUP software. Caffeine (0.248-1.663%) and theobromine (0.106-0.807%) contents were significantly different (p<0.05) depending on the region of origin, with high individual heritability (ĥ²>0.5). The two different progeny groups determined for chlorogenic (1.365-2.281%) and caffeic (0.027-0.037%) acid contents were not significantly different (p<0.05) depending on the locality of origin. Individual heritability values were low to medium for chlorogenic (ĥ²<0.4) and caffeic acid (ĥ²<0.3). The content of the compounds and the values of genetic parameters could support breeding programs for mate.

Download Full-text