Molecular Plant Breeding 2024, Vol.15, No.5, 317-327 http://genbreedpublisher.com/index.php/mpb 319 qSG10Stable, associated with the lignin syringyl-to-guaiacyl ratio, was stable across all tested environments, explaining a significant portion of phenotypic variation (Figure 1) (Zhu et al., 2023). These findings provide an important basis for the genetic improvement and molecular breeding of economic traits in Eucalyptus. Figure 1 QTL mapping in multiple environments (Adopted from Zhu et al., 2023) Image caption: (A) are the genetic distance in cM. (B) are the distribution of all loci in the consensus map. And the distribution of QTLs of growth and wood properties detected in Gonghe (C), Jijia (D) and Yanxi (E) environments. (F) are stable QTLs. (G) are QTL statistics for each trait at each environment (Adopted from Zhu et al., 2023) Another study on Eucalyptus globulus identified five QTLs associated with intumescence severity, demonstrating the genetic basis for this trait and providing a framework for further investigation (Ammitzboll et al., 2018). Additionally, Bayesian mapping in Eucalyptus cladocalyx revealed large-effect pleiotropic QTLs for wood density and slenderness index, highlighting the potential for marker-assisted breeding in low-rainfall environments (Valenzuela et al., 2021). 3.2 Genomic selection and prediction Genomic selection (GS) is an advanced breeding approach that leverages genome-wide marker information to predict the genetic potential of individuals for complex traits (Ballesta et al., 2018). Unlike traditional selection methods, which rely on phenotypic data alone, GS uses dense genetic markers distributed across the genome to build predictive models. These models can accurately estimate the breeding values of individuals, thereby accelerating the selection process and improving the efficiency of breeding programs. Genomic selection (GS) models have shown promise in predicting wood quality traits in Eucalyptus. In Eucalyptus globulus, various GS methods, including Ridge Regression-BLUP (RRBLUP) and supervised principal component regression (PCR), demonstrated high predictive ability for traits such as branching quality and stem straightness (Ballesta et al., 2018). Similarly, in Eucalyptus benthamii, genomic relationship-based models like GBLUP and HBLUP provided more accurate predictions of wood density and tree volume by capturing hidden relatedness and correcting pedigree errors (Paludeto et al., 2021). The inclusion of haplotype effects in Bayesian genomic models further improved the predictive ability for low-heritability traits, suggesting a significant advantage for implementing GS in Eucalyptus breeding programs (Ballesta et al., 2019)
RkJQdWJsaXNoZXIy MjQ4ODYzMg==