Evaluating the accuracy of genomic prediction of growth and wood traits in twoEucalyptusspecies and their F₁hybrids

Abstract

Background: Genomic prediction is a genomics assisted breeding methodology that can increase genetic gains by accelerating the breeding cycle and potentially improving the accuracy of breeding values. In this study, we use 41,304 informative SNPs genotyped in a Eucalyptus breeding population involving 90 E.grandis and 78 E.urophylla parents and their 949 F 1 hybrids to develop genomic prediction models for eight phenotypic traits -basic density and pulp yield, circumference at breast height and height and tree volume scored at age three and six years. We assessed the impact of different genomic prediction methods, the composition and size of the training and validation set and the number and genomic location of SNPs on the predictive ability (PA).

Accuracy of genomic selection for growth and wood quality traits in two control-pollinated progeny trials using exome capture as genotyping platform in Norway spruce

“…When the number of markers increases the computational time for Bayesian method took longer time to converge. Therefore, we also support the proposal that GBLUP is an effective method in providing the best compromise between computational time and prediction efficiency if there are no major gene effects [4,42].…”

“…In our study, heritabilities for wood quality traits were higher than those for tree height, which is expected and agrees with previous reports for Norway spruce [1,28]. Tan et al [4] reported that heritability estimates obtained from GBLUP were higher than those from ABLUP for growth and wood quality traits in Eucalyptus. In contrast, Lenz et al [15] and Gamal EL-Dien et al [12] reported that heritability estimates obtained from GBLUP were lower than those from ABLUP for similar growth and wood quality traits in black spruce and interior spruce (Picea glauca [Moench] Voss  Picea engelmannii Parry ex Engelm.).…”

“…As expected, the accuracies of four different genomic statistical methods did not clearly outperform each other. This contrasts with previous evaluation of the RKHS method that was reported to outperform other GS methods for low heritability growth traits (Tan et al 2017). It was usually observed that different genomic statistical methods produce similar results for growth and wood quality traits in other forest trees species [13,19,41].…”

“…To account for these skewed distributions of a few genes with large effects and most of genes with small effects, Bayes A, B, and Cπ and Bayesian Least Absolute Shrinkage and Selection Operator (BLASSO) were developed to fit the models more accurately, in contrast to Genomic best linear unbiased prediction (GBLUP) model that assumes a normal distribution of allelic effect. In most studies for growth and wood quality traits, the results were similar regardless of models used [4,9]. Resende et al [9] reported that fusiform rust in loblolly pine may be controlled by a few large genes and Bayesian based models had higher predictive ability (PA) defined as the correlation between adjusted phenotype value and GEBV.…”

“…A conventional breeding program for Norway spruce based on pedigree-based phenotypic selection usually takes between 20-30 years in Scandinavian countries [2]. To shorten the breeding cycle, genomic selection (GS) has recently been proposed as an alternative in many tree species such as eucalypts (Eucalyptus) [3][4][5], maritime pine (Pinus pinaster Aiton) [6,7], loblolly pine (Pinus taeda L.) [8,9], white spruce and its hybrids (Picea glauca [Moench] Voss) [10-13] , and black spruce (Picea Mariana [Mill] B.S.P.) [14].…”