Predicting clonal composites performance and enhancing eucalyptus productivity by accounting for indirect genotypic effects

• Ano de defesa: 2023
• Autor(a) principal: Ferreira, Filipe Manoel
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: eng
• Instituição de defesa: Universidade Federal de Viçosa; Genética e Melhoramento
• Palavras-chave em Português: Eucalipto - Melhoramento genético; Genética quantitativa; Modelos multiníveis (Estatísticas); Competição (Biologia); Genética Quantitativa
• Link de acesso: https://locus.ufv.br//handle/123456789/31555; https://doi.org/10.47328/ufvbbt.2023.526

Biotic and abiotic factors have been making it difficult to maintain high rates of realized genetic gain for tree species, especially those cultivated in monocultures. Planting a mixture of genotypes or clonal composites (CC) can be an alternative to increase phytosanitary security and even the productivity of forest plantations. Clones grown in CC may present residual and genetic competition. Competition effects can affect the heritable portion of the total variability and impact the genetic progress of the population under selection. We aim to jointly model the spatial and genetic competition using a linear mixed model at the spatial and genetic level (SCM) to estimate genetic parameters and study the impacts of intergenotypic competition. In addition, we propose a strategy to predict the best combination of clones to compose a CC that has not yet been planted. To the best of our knowledge, no previous study has explored the prediction of CC accounting for competition effects. The main advantage of our methodology consists in modeling the competition at the genetic and residual level to predict the total genotypic value (TGV) of clones and the phenotypic performance of any CC combination. The proposed approach was illustrated in a dataset from clonal trials of eucalyptus in a randomized block design with 24 replications, containing a single tree per plot evaluated for mean annual increment (MAI – m3ha-1ano-1) at ages 3 and 6. The fitted model was efficient in partitioning genetic variation into variations due to direct genotypic effects (DGE) and indirect or competition genotypic effects (IGE). Additionally, we proposed a way to classify clones as aggressive, homeostatic, and sensitive based on the magnitude of the IGE. The SCM was the most suitable according to the Akaike Information Criterion. By accounting for indirect genotypic effects, for MAI, the total heritability decreased from 0.25 to 0.10 for 3 years and from 0.30 to 0.14 for 6 years, compared to a reduced model for IGE. Therefore, heritability was overestimated when IGE was not considered. Based on the TGV, we were able to identify CC with a high expected average performance for MAI, considering the trade-off between DGE and IGE. Therefore, predicting CC by capitalizing on the IGE can provide a strategic advantage in recommending the best combination of clones to be planted. Keywords: Tree Breeding. Quantitative Genetics. Linear Mixed Models. Associative Effects. Competition.