Genomic Predictions With Nonadditive Effects Improved Estimates of Additive Effects and Predictions of Total Genetic Values in Pinus sylvestris

Abstract

Genomic selection study (GS) focusing on nonadditive genetic effects of dominance and the first order of epistatic effects, in a full-sib family population of 695 Scots pine (Pinus sylvestris L.) trees, was undertaken for growth and wood quality traits, using 6,344 single nucleotide polymorphism markers (SNPs) generated by genotyping-by-sequencing (GBS). Genomic marker-based relationship matrices offer more effective modeling of nonadditive genetic effects than pedigree-based models, thus increasing the knowledge on the relevance of dominance and epistatic variation in forest tree breeding. Genomic marker-based models were compared with pedigree-based models showing a considerable dominance and epistatic variation for growth traits. Nonadditive genetic variation of epistatic nature (additive × additive) was detected for growth traits, wood density (DEN), and modulus of elasticity (MOEd) representing between 2.27 and 34.5% of the total phenotypic variance. Including dominance variance in pedigree-based Best Linear Unbiased Prediction (PBLUP) and epistatic variance in genomic-based Best Linear Unbiased Prediction (GBLUP) resulted in decreased narrow-sense heritability and increased broad-sense heritability for growth traits, DEN and MOEd. Higher genetic gains were reached with early GS based on total genetic values, than with conventional pedigree selection for a selection intensity of 1%. This study indicates that nonadditive genetic variance may have a significant role in the variation of selection traits of Scots pine, thus clonal deployment could be an attractive alternative for the species. Additionally, confidence in the role of nonadditive genetic effects in this breeding program should be pursued in the future, using GS.

Increasing genomic prediction accuracy for unphenotyped full-sib families by modeling additive and dominance effects with large datasets in white spruce

“…These contrasting results across studies for acoustic velocity is not unexpected given that the decomposition of genetic variance into additive and dominance components is population specific as it depends on the population allele frequencies ( Falconer and Mackay, 1996 ; Hill et al., 2008 ; Huang and Mackay, 2016 ). In full-sib or clonally replicated trials in other conifers, null or small dominance effects for wood traits were detected in Norway spruce ( Chen et al., 2019 , 2020 ; Nguyen et al., 2022 ) and in Scots pine ( Calleja-Rodriguez et al., 2021 ). In the well-studied Eucalyptus species and their hybrids, the genetic variance of wood density was found to be mostly additive ( Costa e Silva et al., 2004 , 2009 ; Resende et al., 2017 ; Tan et al., 2018 ; Thumma et al., 2022 ).…”

“…Recent studies attempted to decompose additive, dominance, and epistatic (ADE) variances using open-pollinated ( Gamal El-Dien et al., 2016 , 2018 ) or full-sib progeny trials ( Tan et al., 2018 ; Chen et al., 2019 ; Calleja-Rodriguez et al., 2021 ). However, for the studies and traits that showed considerable epistatic variances, it was always associated with large standard errors, and, in all but one case ( Gamal El-Dien et al., 2016 ), GBLUP-ADE was not the best model compared with GBLUP-AD or GBLUP-A based on AIC ( Gamal El-Dien et al., 2018 ; Tan et al., 2018 ; Calleja-Rodriguez et al., 2021 ). Although gene-gene interactions have been found to be pervasive in model organisms ( Mackay, 2014 ), there may be little power to detect epistatic variance for polygenic traits in practice ( Hill et al., 2008 ; Mäki-Tanila and Hill, 2014 ).…”

“…Comparing the performance of GS across studies is difficult because of the different methods used to estimate prediction accuracy in forest trees ( Bousquet et al., 2021 ; Calleja-Rodriguez et al., 2021 ). However, several studies in spruce used similar CV schemes to those used here and calculated the predictive ability ( PA BV ) or accuracy ( PACC BV as ) of breeding values using additive models ( Table 5 ).…”

“…An increasing number of empirical GS studies in forest trees considered non-additive effects and aimed at their separation from additive effects ( Muñoz et al., 2014 ; Bouvet et al., 2016 ; de Almeida Filho et al., 2016 , 2019 ; Gamal El-Dien et al., 2016 , 2018 , 2022 ; Resende et al., 2017 ; Tan et al., 2018 ; Chen et al., 2019 ; Beaulieu et al., 2020 ; Pégard et al., 2020 ; Thavamanikumar et al., 2020 ; Calleja-Rodriguez et al., 2021 ; Thumma et al., 2022 ). Compared with ABLUP, GBLUP was shown to better separate both types of variance and reduce confounding between genetic and environmental effects ( Muñoz et al., 2014 ; Gamal El-Dien et al., 2016 ; Tan et al., 2018 ).…”

Pervasive Under-Dominance in Gene Expression Underlying Emergent Growth Trajectories inArabidopsis thalianaHybrids

“…However, incorporating non-additive, i.e.,, dominant and epistatic, effects into quantitative genetics modeling can improve heritability estimates and accuracy of genomic prediction (Su et al 2012;Wellmann and Bennewitz 2012;Kumar et al 2015;Xiang et al 2016). Limited empirical data exists for non-additive variance estimates, which varies greatly between organisms and can range from around 3-15% of total phenotypic variance in humans and animals (Misztal 1997;Zhu et al 2015) to a third or more in plants (Kumar et al 2015;Berguson et al 2017;Calleja-Rodriguez et al 2021). Adequate knowledge of non-additive genetic action is therefore of pivotal importance for a thorough understanding of the genetic architecture of complex traits, especially those that are fitness-related, and their evolutionary trajectory.…”